Journal of Personality and Social Psychology Risky Business: When Humor Increases and Decreases Status T. Bradford Bitterly, Alison Wood Brooks, and Maurice E. Schweitzer Online First Publication, November 10, 2016. http://dx.doi.org/10.1037/pspi0000079

CITATION Bitterly, T. B., Brooks, A. W., & Schweitzer, M. E. (2016, November 10). Risky Business: When Humor Increases and Decreases Status. Journal of Personality and Social Psychology. Advance online publication. http://dx.doi.org/10.1037/pspi0000079

Journal of Personality and Social Psychology 2016, Vol. 111, No. 10, 000

© 2016 American Psychological Association 0022-3514/16/$12.00 http://dx.doi.org/10.1037/pspi0000079

Risky Business: When Humor Increases and Decreases Status T. Bradford Bitterly

Alison Wood Brooks

Wharton School

Harvard Business School

Maurice E. Schweitzer

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Wharton School Across 8 experiments, we demonstrate that humor can influence status, but attempting to use humor is risky. The successful use of humor can increase status in both new and existing relationships, but unsuccessful humor attempts (e.g., inappropriate jokes) can harm status. The relationship between the successful use of humor and status is mediated by perceptions of confidence and competence. The successful use of humor signals confidence and competence, which in turn increases the joke teller’s status. Interestingly, telling both appropriate and inappropriate jokes, regardless of the outcome, signals confidence. Although signaling confidence typically increases status and power, telling inappropriate jokes signals low competence and the combined effect of high confidence and low competence harms status. Rather than conceptualizing humor as a frivolous or ancillary behavior, we argue that humor plays a fundamental role in shaping interpersonal perceptions and hierarchies within groups. Keywords: competence, confidence, humor, interpersonal perception, status

Costolo’s experience suggests that humor can help an individual climb the corporate ladder, but Sacco’s experience offers a cautionary tale of the inherent risks in using humor. We postulate that humor can profoundly influence status, and we argue that humor is a pervasive but underinvestigated behavioral construct. Across eight studies, we investigate how the use of humor influences status. We conceptualize the use of humor as a risky behavior, and we explore how the appropriateness of humor and observers’ reactions to humor attempts (e.g., laughter) influence whether the joke teller’s status increases or decreases.

Dick Costolo, the former CEO of Twitter, began his career in improvisational comedy, and he attributes much of his success in business to his use of humor (Bilton, 2012). The night before Costolo joined Twitter as Chief Operating Officer in September 2009, he tweeted: “First full day as Twitter COO tomorrow. Task #1: undermine CEO, consolidate power.” (Costolo, 2009). A year later, he became the Chief Executive Officer. Just as humor can contribute to career success, it can lead to tumultuous falls. On December 20, 2013, before leaving Heathrow Airport for South Africa, Justine Sacco, a Public Relations Representative for IAC, a media and Internet company, tweeted: “Going to Africa. Hope I don’t get AIDS. Just kidding. I’m white!” (Ronson, 2015). Her humor attempt provoked a firestorm of criticism, and ultimately cost Sacco her job.

Status Status is ubiquitous and consequential. Across cultures, across organizations, and across social hierarchies, individuals are highly motivated to achieve greater status (Anderson, Hildreth, & Howland, 2015; Anderson, John, Keltner, & Kring, 2001; Barkow et al., 1975; Maslow, 1943). Status is the relative level of respect, prominence, and esteem that an individual possesses within a dyad or group (Anderson, Hildreth, & Howland, 2015; Anderson & Kilduff, 2009a; Anderson, Willer, Kilduff, & Brown, 2012; Berger, Cohen, & Zelditch, 1972; Flynn, Reagans, Amanatullah, & Ames, 2006; Goldhamer & Shils, 1939; Kilduff & Galinsky, 2013; Magee & Galinsky, 2008; Pettit & Lount, 2010). Status is a defining characteristic of human interaction; every social group has a status hierarchy (Anderson, Srivastava, Beer, Spataro, & Chatman, 2006; Berger et al., 1972; Magee & Galinsky, 2008; Mazur, 1973; Ridgeway, 1987). In addition to being ubiquitous, status is important. Compared with low-status individuals, high-status individuals have greater access to resources (e.g., money, social support), and enjoy greater physical and psychological well-being (Adler, Epel, Castellazzo, & Ickovics, 2000; Anderson & Kilduff, 2009a; Ellis, 1994; Marmot, 2004; Pettit & Sivanathan, 2012). The allure of obtaining

T. Bradford Bitterly, Department of Operations, Information & Decisions, Wharton School; Alison Wood Brooks, Negotiation, Organizations & Markets Unit, Harvard Business School; Maurice E. Schweitzer, Department of Operations, Information & Decisions, Wharton School. The authors are grateful for advice and feedback from Katherine Milkman, Emma Levine, Ellen Evers, Jessica Kennedy, Eric VanEpps, and members of the OID-Decision Processes Lab. We thank Cameron Anderson for guidance and suggestions. The authors are appreciative of the valuable research assistance from Graham Overton and Liam Murphy. The authors thank the Wharton Behavioral Lab and the Wharton Risk Center Russell Ackoff Doctoral Student Fellowship for financial support. We thank Joshua Lewis for allowing us to formally launch his modeling career with the photos depicted in Appendix E. Correspondence concerning this article should be addressed to T. Bradford Bitterly, Operations, Information & Decisions, Wharton School, 3730 Walnut Street, 526.1 Jon M. Huntsman Hall, Philadelphia, PA 19104. E-mail: [email protected] 1

BITTERLY, BROOKS, AND SCHWEITZER

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

2

higher status is strong (Anderson, Hildreth, & Howland, 2015; Anderson et al., 2001; Barkow et al., 1975; Frank, 1985; Hardy & Van Vugt, 2006; Loch, Huberman, & Stout, 2000; Maslow, 1943; Pettit & Sivanathan, 2011; Rucker & Galinsky, 2008; Sivanathan & Pettit, 2010; Willer, 2009). To gain status, individuals endeavor to display competence. Groups accord greater respect and influence to individuals who demonstrate superior abilities (Anderson & Kilduff, 2009a; Berger et al., 1972; Lord, De Vader, & Alliger, 1986). In many cases, however, individuals lack objective information about how competent an individual is, and rely on signals instead. As a result, behaviors that signal competence increase status (Anderson, Brion, Moore, & Kennedy, 2012; Anderson & Kilduff, 2009a, 2009b; Chen, Peterson, Phillips, Podolny, & Ridgeway, 2012; Kennedy, Anderson, & Moore, 2013). For example, in a new encounter, individuals who express overconfidence and act in a domineering way can signal competence and boost their status (Anderson et al., 2012; Anderson & Kilduff, 2009b; Kennedy, Anderson, & Moore, 2013). That is, by appearing competent (e.g., projecting confidence, sharing good ideas, making intelligent comments), individuals can increase their status. In our investigation, we examine an unexplored method by which individuals might signal competence and increase their status: using humor.

ence members and was not benign enough to be perceived as funny. In the current research, we explore how humor attempts influence the perceived competence and confidence of a joke teller. Humor is risky; an expresser’s humor attempt can fall flat in different ways. First, if the target or audience perceives the humor attempt to be merely benign, it might not be obvious that the expresser was attempting to use humor at all. Second, if the humor attempt is not interesting, exciting, or entertaining, then the target or audience may view the humor attempt as boring. Third, a humor attempt may fail by offending the joke target, the audience, or both. As Justine Sacco learned, it is easy to offend others, especially because humor norms vary across contexts and individuals (Daniel, O’Brien, McCabe, & Quinter, 1985; Feingold, 1992; Martin, 2007; McGraw & Warner, 2014; Smeltzer & Leap, 1988; Thomas & Esses, 2004). Prior to attempting to use humor, the expresser cannot be certain of how the audience will react. This is particularly true when the joke teller is interacting with an unfamiliar audience; the joke teller cannot be certain of what the audience views as acceptable, and the audience does not know the intentions behind the teller’s comment. The act of attempting to use humor demonstrates confidence because humor attempts may fall flat or offend the audience. We expect observers to infer this and evaluate individuals who attempt to use humor as more confident than those who do not.

Humor Consistent with prior work, we define humor as an event between two or more individuals in which at least one individual experiences amusement and appraises the event as funny (adapted from Cooper, 2005, 2008; Gervais & Wilson, 2005; Martin, 2007; McGraw & Warren, 2010; McGraw, Warren, & Kan, 2015; Warren & McGraw, 2015, 2016). We define a joke as a humor attempt, and we conceptualize humorous encounters as interactions between three focal actors: the expresser, the target(s), and the audience. Targets of jokes can be specific or general, and human, nonhuman, or even inanimate. In some cases, the target and audience are the same (e.g., teasing), or the expresser and the target are the same (e.g., self-deprecating humor). In a humor attempt, the expresser acts with the intention to amuse and elicit mirth from the audience. Importantly, humor attempts may or may not be successful. When an expresser attempts to use humor, observers will judge the success of the humor attempt based on several factors, including, but not limited to, the appropriateness of the humor attempt and whether or not the attempt elicits laughter. Prior work suggests that humor is successful when someone perceives the attempt to be a benign violation (McGraw & Warren, 2010; Veatch, 1998; Warren & McGraw, 2016, 2015). That is, for a humor attempt to be perceived as funny, it must be two things. First, it must violate physical or psychological safety (e.g., violations of linguistic, social, or moral norms). Second, it must be benign. For example, Dick Costolo violated social norms by tweeting that he intended to undermine his CEO. However, the norm violation was not overtly offensive—it was clear that he was not actually attacking his CEO. Similarly, Justine Sacco violated social norms by tweeting that she could not get AIDS because she is white. However, by joking about the correlation between a devastating illness and race, Justine Sacco’s humor attempt was deemed offensive by many audi-

Hypothesis 1: The use of humor increases perceptions of confidence. The willingness to use humor signals confidence, but it is the successful use of humor that signals competence. The successful use of humor requires the expresser to recognize the opportunity to say something funny and deliver the joke, while navigating the risks of being either boring or offensive. A substantial literature has documented a close association between the successful use of humor and competence. For example, in a study of children (ages 10 –14 years old), Masten (1986) found a correlation between the successful use of humor and IQ, and a correlation between the successful use of humor and school performance. In addition, Masten (1986) found that kids who used humor successfully were liked better by their teachers and their peers. In related work, Decker (1987) found that employees’ ratings of their supervisor’s sense of humor correlated with ratings of the supervisor’s intelligence, confidence, and effectiveness. The link between humor and competence has also been established with abstract reasoning and verbal intelligence tests (Greengross & Miller, 2011). Though correlational, these studies suggest that the use of humor is closely associated with competence. Humor has also been linked with performance and creativity (Huang, Gino, & Galinsky, 2015; Lehmann-Willenbrock & Allen, 2014; Martin, 2007). Lehmann-Willenbrock and Allen (2014) found that group performance was positively associated with the use of humor. Humorous remarks that were positive, not mean or disparaging, and successful were associated with greater functional communication behaviors (e.g., procedural statements such as goal-oriented statements and socioemotional statements such as encouragement). Importantly, Lehmann-Willenbrock and Allen (2014) found that the successful use of humor prompted novel idea generation. In a series of experiments, Huang, Gino, and Galinsky

HUMOR AND STATUS

(2015) found that individuals who express sarcasm perform better on creativity tasks. Sarcasm is a form of humor in which an individual communicates a message using words that mean the opposite of the literal statement (Gibbs, 1986; Huang, Gino, & Galinsky, 2015; Pexman & Olineck, 2002). We postulate that the association between humor and competence is pervasive and familiar, and that people will make the inference that those who effectively use humor are competent.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Hypothesis 2: The successful use of humor increases perceptions of competence. Signaling greater confidence and competence can boost status (Anderson et al., 2012; Anderson & Kilduff, 2009a, 2009b; Chen et al., 2012; Kennedy, Anderson, & Moore, 2013; Ridgeway, 1991). Consequently, we predict that the effective use of humor can increase status by signaling confidence and competence. Specifically, we expect perceptions of confidence and competence to mediate the relationship between humor and status. Hypothesis 3: The successful use of humor increases status. Hypothesis 4a: Perceptions of confidence mediate the relationship between the use of humor and status. Hypothesis 4b: Perceptions of competence mediate the relationship between the use of humor and status. A few studies have linked the successful use of humor with influence in interpersonal settings (Avolio, Howell, & Sosik, 1999; Lehmann-Willenbrock & Allen, 2014; O’Quin & Aronoff, 1981). The use of humor can increase concession-making in negotiations (Kurtzberg, Naquin, & Belkin, 2009; O’Quin & Aronoff, 1981), and Avolio, Howell, and Sosik (1999) found that leaders in productive groups were more likely to use humor successfully than were those in unproductive groups. Lehmann-Willenbrock and Allen (2014) identify a link between the use of humor and performance. This relationship, however, only existed in certain instances, such as when a joke was followed by either laughter or another joke. The successful use of humor may increase influence by boosting positive affect. Increased positive affect has been shown to increase positive evaluations of others and draw attention away from negative information (Lyttle, 2001; Strick, Holland, van Baaren, & van Knippenberg, 2012). On the other hand, a humor attempt that does not succeed because it is offensive (e.g., Justine Sacco’s tweet) might induce negative affect, which could harm the audience’s evaluations of the joke teller. Although prior work has linked humor with interpersonal influence and established that leadership requires the ability to influence others (Yukl, Wall, & Lepsinger, 1990), no prior work has conceptualized humor as a tool for gaining status. Surprisingly, prior humor research has focused almost exclusively on the successful use of humor. In practice, many humor attempts fail because they are too benign, boring, or inappropriate. Forecasting appropriateness is difficult, because the appropriateness of humor is highly context dependent (Campos, Keltner, Beck, Gonzaga, & John, 2007; Hoption, Barling, & Turner, 2013; Keltner, Capps, Kring, Young, & Heerey, 2001; Keltner, Young, Heerey, Oemig, & Monarch, 1998; Martin, 2007; McGraw, Warren, Williams, & Leonard, 2012; Lyttle, 2001; Robert, Dunne, &

3

Iun, 2016). In our research, we conceptualize the appropriateness of a humor attempt to reflect both the type of joke told (e.g., self-deprecation, puns, insults, sexual innuendos) and the fit of that joke in context. We consider the appropriateness of a humor attempt as a moderator of the relationship between humor and status. We expect the use of appropriate humor to be more successful in boosting status than the use of inappropriate humor. Attempting to use both appropriate and inappropriate humor requires confidence, and demonstrating confidence is typically associated with competence and higher status (Anderson et al., 2012; Anderson & Kilduff, 2009a, 2009b; Chen et al., 2012; Kennedy, Anderson, & Moore, 2013; Ridgeway, 1991). However, by making inappropriate jokes, expressers signal that they are ignorant of social boundaries, that they have failed to understand and follow norms (e.g., making racist, sexist, or otherwise bigoted remarks), and that they lack competence. Although an individual who tells an inappropriate joke may signal confidence to the audience, the audience also receives a signal of ignorance. As a result, in contrast to the use of appropriate humor, the use of inappropriate humor can demonstrate confidence, but can signal a lack of competence and lower status. Hypothesis 5: Appropriateness of the humor attempt will moderate the relationship between humor and competence. Public reactions to humor attempts can profoundly shape perceptions of the humor attempt. For example, individuals are more likely to laugh when they hear others laugh (Provine, 1992; Olson, 1992; Smyth & Fuller, 1972), and laughter from the target and/or audience serves as a public demonstration that the expresser’s humor attempt was successful. Laughter demonstrates amusement and approval (Sauter, Eisner, Ekman, & Scott, 2010), and because people pay more attention to individuals whom others approve of (Chudek, Heller, Birch, & Henrich, 2012), we expect humor attempts that elicit laughter to be more effective in boosting status than humor attempts that fail to elicit laughter. In contrast to a humor attempt that elicits laughter, a humor attempt that fails to elicit laughter signals low competence. Hypothesis 6: Laughter will moderate the relationship between humor and competence. Taken together, we summarize our theoretical framework (Hypotheses 1– 6) in Figure 1. Our research program advances our

Figure 1. Theoretical model.

BITTERLY, BROOKS, AND SCHWEITZER

4

theoretical and practical understanding of humor and status. We are the first to explore how humor attempts influence status. In contrast to prior humor research that has focused on successful humor attempts, we consider the consequences of both successful and unsuccessful humor attempts. In exploring unsuccessful humor attempts, we consider jokes that fail to elicit laughter and jokes that are perceived as inappropriate.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Overview of Current Work Our work investigates the relationship between humor attempts and status. Though humor attempts can involve nonverbal expressions, in our investigation, we operationalize humor attempts using spoken jokes. We motivate our investigation with two pilot studies. In these pilot studies, we identify workplace humor as a common phenomenon, and we pilot test each of the jokes we use in Studies 1 through 4 to gauge how funny and appropriate they are. In Studies 1a, 1b, 2a, and 2b, we explore the relationship between humor and status using different contexts and different jokes. We also investigate how humor changes perceptions of confidence and competence, our proposed mediating mechanisms. In Studies 2 through 4, we examine the moderating role of joke success as signaled by audience laughter and joke appropriateness. In Studies 2a and 2b, we consider the moderating role of audience laughter. In Studies 3a, 3b, 4a, and 4b, we consider the moderating role of joke appropriateness. We test the moderating roles of laughter and appropriateness in an organizational setting because the workplace provides a context where there are higher standards for professional behavior, and norms of appropriateness matter (e.g., a joke with sexual content is typically seen as inappropriate for a professional setting but might be viewed as acceptable in a more casual setting outside of work). Though successful humor attempts are likely to increase perceptions of confidence, competence, and status, humor attempts may harm perceptions of competence and status when a joke is perceived to be inappropriate or when the audience fails to laugh.

Pilot Study 1: Pervasiveness of Humor at Work In Pilot Study 1, we recruited 200 working adults to investigate the pervasiveness of humor in the workplace and to motivate our investigation of humor in organizations.

Method Participants. We recruited 200 participants (118 male, 82 female) online via Amazon.com’s Mechanical Turk to participate in a survey in exchange for $0.40. Participants were, on average, 29.4 years old (SD ⫽ 8.58), and 100% were partially or fully employed at the time of the survey. Design and procedure. First, we asked participants to recall a joke a coworker had told in the past and when it was told. We also asked participants to indicate their agreement with statements that their coworkers frequently made jokes and that it would be normal for jokes like the one they recalled to be told in the workplace (1 ⫽ strongly disagree; 7 ⫽ strongly agree). Finally, participants reported their demographic information (age, gender). Next, we had three research assistants rate the extent to which they agreed with the following statement: “The joke is appropriate

for a coworker to tell to another coworker” (1 ⫽ strongly disagree, 7 ⫽ strongly agree). The ratings across research assistants were consistent (␣ ⫽ .90).

Results and Discussion Results from this study reveal that telling jokes is a common workplace behavior. Only one participant (0.5% of our sample) was not able to recall a joke, and 74% of the recalled jokes had been heard within the past month. Participants reported that coworkers other than the joke-teller in their example make similar jokes (M ⫽ 4.99, SD ⫽ 1.34), and that their coworkers frequently make jokes (M ⫽ 5.40, SD ⫽ 1.17). We also find that both appropriate and inappropriate jokes are common in the workplace (appropriateness rating M ⫽ 4.94, SD ⫽ 1.90, 27% ⬍ 4, 70% ⬎ 4). All jokes from this study are available from the corresponding author upon request.

Pilot Study 2: Testing Joke Funniness and Appropriateness Across Studies 1 through 4, we use nine different jokes. In this pilot study, we assess the funniness, boringness, and appropriateness of each joke. We use variance in the ratings of appropriateness of these jokes to test our hypotheses.

Method Participants. We recruited 457 participants (264 male, 193 female) online via Amazon.com’s Mechanical Turk to participate in a short study in exchange for $0.20. The participants were, on average, 32.96 years old (SD ⫽ 10.97). Design and procedure. Each participant evaluated one of the nine jokes listed in Table 1 and described in Appendix A. We presented participants with joke scenarios that depict either a customer testimonial or a meeting between a manager and a job candidate. The scenarios end after the joke and did not include information about how other individuals reacted to the joke. After reading one of the nine joke scenarios, participants rated the last comment made by the customer/candidate on eight dimensions (funny, humorous, boring, dull, inappropriate, appropriate, tasteless, suitable). The response scale ranged from 1 (not at all) to 7 (extremely). We combined “funny” and “humorous” to form a rating of funniness for each joke (r ⫽ .93). We combined “boring” and “dull” to form a rating of boringness for each joke (r ⫽ .80), and we combined “appropriate” and “suitable” with reverse scores for “inappropriate” and “tasteless” to form a rating of appropriateness for each joke (␣ ⫽ .92). Finally, we asked participants demographic questions (age, gender).

Results and Discussion Participant ratings of funniness were moderate to high across all nine joke scenarios (all means were above 3.39), ratings of boringness were low (all means were below 2.99), and—as intended—ratings of appropriateness varied across the nine jokes. The varied appropriateness of these jokes enabled us to investigate the effects of joke appropriateness on interpersonal perception. Confirming our expectations in selecting these jokes, the appropriateness ratings were significantly lower for the “inappropriate” jokes

HUMOR AND STATUS

5

Table 1 Ratings of Joke Funniness, Boringness, and Appropriateness (Pilot Study 2) Study

Joke

Funny M (SD)

Boring M (SD)

1A 1B 2B 3A 3B 4A

Pet Waste Swiss Flag Five-Year Anniversary That’s What She Said Receptionist When I Die Two Chicks CrossFit Goodyear

4.56abc (1.77) 3.53cd (1.95) 4.83ab (1.55) 3.74bcd (2.04) 3.40d (2.02) 4.31abcd (1.59) 3.97abcd (2.05) 4.94a (1.53) 4.97a (1.55)

2.04a (1.31) 2.98b (1.75) 2.30ab (1.35) 2.80ab (1.75) 2.11ab (1.24) 2.79ab (1.37) 2.46ab (1.73) 2.14ab (1.15) 2.01a (1.34)

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

4B

Appropriate M (SD) 4.26a (1.43) 5.50b (1.02) 4.36a (1.47) 1.93cd (0.94) 1.79d (1.21) 4.40a (1.36) 1.80d (0.81) 4.86ab (1.40) 2.62c (1.35)

Note. Mean funniness, boringness, and appropriateness ratings for the jokes used in each study. Means in each column with different subscripts are significantly different at p ⬍ .05 level in pairwise t tests using a Bonferroni correction.

we would use in Study 3a (M ⫽ 1.93, SD ⫽ 0.94), Study 3b (M ⫽ 1.79, SD ⫽ 1.21), Study 4a (M ⫽ 1.80, SD ⫽ 0.81), and Study 4b (M ⫽ 2.62, SD ⫽ 1.35) than the appropriateness ratings for the other jokes we used in our studies. We present these results in Table 1.

Study 1: Successful Humor Increases Status In Study 1, we investigate the influence of humor on status. In Study 1a, participants rated the status of a presenter who either attempted or did not attempt to use humor in a face-to-face interaction. In Study 1b, participants nominated individuals who either did or did not attempt to use humor in a face-to-face interaction as leaders for a subsequent task.

Study 1a Method Participants. We recruited 166 adults from a city in the northeastern United States to participate in a behavioral lab study in exchange for $10. A total of 160 people completed the study (66 male, 94 female, Mage ⫽ 24.86 years, SD ⫽ 9.39). Design and procedure. After checking into the behavioral lab, participants (along with two confederates who also checked into the behavioral lab) walked to a nearby classroom where they completed the study. The largest group had 15 people (13 participants and two confederates); the smallest group had six people (four participants and two confederates). In the classroom, we sat each participant at their own desk with a packet of materials. As participants read the materials, we asked them to imagine that they were writing customer testimonials for a pet waste removal service, FastScoop.com. We informed them that FastScoop was running a contest, looking for customer testimonials, with the hope that the testimonials would attract attention for the service. We then presented participants with a background photo for an advertisement for FastScoop and asked them to write a brief (1–3 sentence) testimonial to accompany the photo. We include an advertisement photo very similar to the one used in the study in Appendix B (the original is available upon request from the authors). We gave participants three minutes to write their testimonials.

We told participants that each of them would present their testimonials in front of the rest of the participants in a randomly determined order. After completing their testimonials, we asked participants to draw a number from an envelope to determine the order in which they would present. The envelope contained pieces of paper numbered 3 to 25. We omitted the numbers one and two from the envelope, so that the two confederates would always present first and second. We used a mixed between- and within-subjects design, in which one confederate delivered a serious testimonial, and the other confederate alternated between presenting a humorous and a serious testimonial by lab session. Across all 16 sessions, we used the same two male confederates who switched presenting either first or second each day. After all participants drew a number, the experimenter asked the participant who had drawn the number 1 to come to the front of the room and present his testimonial in front of the group. The first confederate placed his testimonial on a document camera, which projected the testimonial on a screen in front of the room. The first testimonial was always a serious testimonial, which set the tone and expectation for the exercise. The confederate projected their handwritten testimonial on the screen and read their testimonial out loud. The serious testimonial read, “They come every week and are very dependable! Overall, a great waste removal service!” Next, the experimenter asked the participant who had drawn the number 2 to come to the front of the room and present his testimonial. Half of the time, the second confederate delivered a humorous testimonial, and half of the time the second confederate delivered a serious testimonial. We alternated the treatment condition each laboratory session. The humorous testimonial read, “Very professional. After cleaning up the poop, they weren’t even upset when they found out that I don’t have a pet! But seriously, this service is reliable and always leaves the yard spotless!” The serious testimonial that the second confederate delivered was, “Very professional. This service is reliable and always leaves the yard spotless!” After each confederate delivered their testimonial, we asked participants to complete a customer testimonial evaluation form. The testimonial evaluation form asked participants to rate the presenter’s customer testimonial, using a 7-point scale (1 ⫽ not at all, 7 ⫽ extremely), on the following qualities: engaging, funny, appropriate, entertaining, succinct, clear, memorable, and effective. Ratings of funniness served as our manipulation check. We were also interested

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

6

BITTERLY, BROOKS, AND SCHWEITZER

in participant ratings of the appropriateness of the testimonial. We included the other items to mask the purpose of the study. We also asked participants to rate, using the same 7-point scale, other characteristics about the confederates: independent, powerful, low status, respected, competent, confident, intelligent, capable, and skillful. We combined the first four items evaluating the presenter (with low status reverse-coded) to form an index of status conferral (adapted from Brescoll & Uhlmann, 2008; Tiedens, 2001; ␣ ⫽ .64), our main dependent variable.1 We used the “confident” item to measure confidence. We combined the remaining four items to form a competence index (adapted from Fiske, Cuddy, Glick, & Xu, 2002; ␣ ⫽ .92). After participants rated the second confederate, the experimenter announced that due to time constraints, no additional participants would present. The experimenter then handed out the exit questionnaire, which asked participants to provide their age, gender, and any additional comments.

Results and Discussion Manipulation check. Our manipulation checks confirmed that our humor manipulation was successful. Participants rated the humorous testimonial (M ⫽ 6.13, SD ⫽ .99) as significantly funnier than the serious testimonial (M ⫽ 2.31, SD ⫽ 1.41), t(15) ⫽ 29.26, p ⬍ .0001. We found that participants viewed the humorous testimonial (M ⫽ 5.20, SD ⫽ 1.36) as less appropriate than the serious testimonial (M ⫽ 5.71, SD ⫽ 1.27), t(15) ⫽ 3.11, p ⬍ .01. However, ratings of the appropriateness of the humorous testimonial were well above the midpoint of the scale. Although participants viewed the humorous testimonial as less appropriate than the serious one, they did not view the humorous testimonial as inappropriate. None of the experimental control variables (research assistant that presented, age, and sex of the participant) influenced how funny or appropriate participants rated the humorous and serious testimonials of the second presenter. Main results. We report our results controlling for confederate fixed effects, ratings of the first presenter, and clustering standard errors by session.2 Status. The status of the second presenter was significantly higher when he delivered the humorous testimonial (M ⫽ 5.03, SD ⫽ 0.76) than when he delivered a serious testimonial (M ⫽ 4.43, SD ⫽ 0.89), t(15) ⫽ 5.95, p ⬍ .0001, 95% CI [0.38, 0.80], simulated power ⫽ .99 at an ␣ of .05 using 1000 simulations (not clustered by session without fixed effects). Male participants rated the second presenter as higher on status than female participants did (p ⬍ .05). None of the remaining experimental control variables (which confederate delivered the second testimonial or participant age) influenced status. Competence. Ratings of competence of the second presenter were also significantly higher when he delivered a humorous testimonial (M ⫽ 5.32, SD ⫽ 0.93) than when he delivered a serious testimonial (M ⫽ 4.90, SD ⫽ 0.99), t(15) ⫽ 4.00, p ⬍ .01, 95% CI [0.28, 0.91]. None of the experimental control variables (which confederate delivered the second testimonial, participant gender, or participant age) influenced perceptions of competence. We depict these results in Figure 2. Confidence. We find that the second presenter was rated as significantly more confident when he delivered a humorous testimonial (M ⫽ 5.64, SD ⫽ 1.07) than when he delivered a serious

testimonial (M ⫽ 4.70, SD ⫽ 1.23), t(15) ⫽ 6.46, p ⬍ .0001, 95% CI [0.71, 1.41]. Mediation. Both competence and confidence mediated the relationship between the second presenter’s testimonial (humorous vs. serious) and status. This is true across both Baron and Kenny (1986) and bootstrap analyses (Preacher & Hayes, 2004, 2008). We provide details of the mediation analyses for every study in Appendix F and summarize the mediation analysis in Table 5. Summary. In Study 1a, we found that when an individual makes a comment that is funny and appropriate, others view him as higher in confidence and competence, which leads to higher ratings of status. Increased ratings of confidence and competence mediated the relationship between the use of humor and judgments of status. We summarize the status findings across all of our studies in Table 2.

Study 1b We extend our investigation of humor and status in Study 1b by using a different joke, a different attitudinal measure of status, and a behavioral measure of status.

Method Participants. We recruited 210 adults from a city in the northeastern United States to participate in a behavioral lab study in exchange for $10. A total of 190 people completed the study (32.8% male, Mage ⫽ 19.94 years, SD ⫽ 1.70). The modal session included 13 participants and 2 confederates. Across the 15 sessions, the number of participants per session ranged from 9 to 13. Design and procedure. The procedure for Study 1b was largely the same as Study 1a, with three notable changes. First, we used a different context with a different joke. Second, we used a different attitudinal measure of status, and third, we included a behavioral measure of status. Scenario and joke. We asked participants to imagine that they were writing customer testimonials for a hypothetical travel service, VisitSwitzerland.ch. We informed them that VisitSwitzerland was soliciting customer testimonials for a competition, hoping to attract attention for their travel service. We then presented participants with a photo for an advertisement for VisitSwitzerland. We include an advertisement photo very similar to the one used in the study in Appendix C (the original is available upon request from the authors). Note that the photo of Switzerland includes Switzerland’s flag (a red background with a white cross). We then gave participants 3 minutes to write a brief (1–2 sentence) 1 We also ran our analysis without “independent” in our index of status conferral. We find that excluding “independent” does not change our results. This is true for all studies where we use this index of status conferral (Studies 2b, 3a, 3b, and 4a). 2 We control for confederate fixed effects to account for any results which are driven by the research assistant that was delivering the second testimonial. We control for participant ratings of the first presenter to account for participant level differences in ratings. We cluster the standard errors by session, because randomization occurred at the session level and participant reactions within each session are not independent. The results are unchanged if we do not control for confederate fixed effects, cluster by session, and control for ratings of the first presenter (ps ⬍ .01 for Funniness, Status, Competence, and Confidence; p ⬍ .05 for Appropriateness).

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

HUMOR AND STATUS

Figure 2.

7

Joke teller status, competence, and confidence in Study 1a: Pet Waste Testimonial.

testimonial to accompany the advertisement to answer the question, “What made you fall in love with Switzerland?” As in Study 1a, the first confederate always presented a serious testimonial. The first testimonial read, “The country is beautiful. The scenery is truly breathtaking!” Half of the time, the second confederate delivered a humorous testimonial, and half of the time the second confederate delivered a serious testimonial. We alternated the treatment condition each laboratory session. In the humor condition, the testimonial included a joke, “The mountains are great for skiing and hiking, and the flag is a big plus! Seriously, it’s amazing!” In the serious condition, the testimonial read, “The mountains are great for skiing and hiking! It’s amazing!” Attitudinal measures. As in Study 1a, after each confederate delivered their testimonial, we asked participants to complete a customer testimonial evaluation form. Using 7-point scales (1 ⫽ not at all, 7 ⫽ extremely), participants rated the testimonials on the following qualities: engaging, funny, appropriate, entertaining, succinct, clear, memorable, and effective. Ratings of funniness served as our manipulation check. We were also interested in participant ratings of the appropriateness of the testimonial. We included the other items to mask the purpose of the study.

We asked participants to rate other characteristics about the confederates: respected, admired, influential, competent, confident, intelligent, capable, and skillful (7-point scales). We used the first three items to measure status (adapted from Anderson, Kraus, Galinsky, & Keltner, 2012; Kilduff & Galinsky, 2013; ␣ ⫽ .88), the item “confident” to measure confidence, and the remaining four items to measure competence (adapted from Fiske, Cuddy, Glick, & Xu, 2002; ␣ ⫽ .92). Behavioral measure. After participants rated the second confederate, the experimenter announced that due to time constraints, no additional participants would present. The experimenter then asked participants to complete a Group Leader form, our behavioral measure of status. The form instructed participants that later in the lab session we would ask them to complete a group task. We informed participants that the groups would be randomly determined, and any of the other participants could be assigned to their group. Each participant had 25 points to allocate to each presenter or themselves, based on the extent to which they would like that individual to be the leader of their group. We used the number of points the participants gave to each presenter as our behavioral measure of status (adapted from Halevy, Chou, Cohen, & Livingston, 2012).

Table 2 Ratings of Status Across All Studies Study

Joke

Serious M (SD)

1A 1B 2A 2B 3A 3B 4A

Pet Waste Swiss Flag Coworker Five-Year Anniversary That’s What She Said Receptionist When I Die Two Chicks CrossFit Goodyear

4.43a (0.89) 4.85a (4.84) 4.18a (1.40) 4.15a (1.08) 4.58a (1.08) 4.20a (1.00) 4.07a (0.84) 4.07a (0.84) 3.94a (1.27) 3.94a (1.27)

4B

Failed Joke M (SD)

Successful Joke M (SD)

3.94a (1.46) 4.22a (1.19) 3.30b (0.93) 3.27b (0.98) 3.68a (0.76) 2.97b (0.99) 3.27b (1.27) 2.57b (1.33)

5.03b (0.76) 6.66b (6.32) 4.94b (1.18) 4.95b (0.83) 3.86c (0.96) 3.97a (1.29) 4.62b (0.88) 3.73a (0.93) 4.84c (0.95) 3.61a (1.50)

Note. Means in each column with different subscripts are significantly different at p ⬍ .05 level. For Study 1b, we present the leadership points allocated to the second presenter based on condition.

BITTERLY, BROOKS, AND SCHWEITZER

8

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Results and Discussion Manipulation check. The manipulation checks confirmed that our humor manipulation was successful. Participants rated the humorous testimonial (M ⫽ 4.53, SD ⫽ 1.88) as significantly funnier than the serious testimonial (M ⫽ 2.16, SD ⫽ 1.42), t(15) ⫽ 8.22, p ⬍ .001. We next considered ratings of the appropriateness of the two testimonials presented second. We found no significant differences in appropriateness ratings between the humorous testimonial (M ⫽ 5.57, SD ⫽ 1.25) and the serious testimonial (M ⫽ 5.52, SD ⫽ 1.28), t(15) ⫽ 0.95, p ⫽ .36. None of the experimental control variables (research assistant that presented, age, and sex of the participant) influenced how funny or appropriate participants rated the humorous and serious testimonials of the second presenter. Main results. As in Study 1a, we report all results controlling for confederate fixed effects, ratings of the first presenter, and clustering standard errors by session. Status. The number of leader points allocated to the second presenter was significantly higher when the confederate delivered the humorous testimonial (M ⫽ 6.66, SD ⫽ 6.32) than when the confederate delivered the serious testimonial (M ⫽ 4.85, SD ⫽ 4.84), t(15) ⫽ 3.13, p ⬍ .01, 95% CI [0.52, 2.75], simulated power ⫽ .58 at an ␣ of .05 using 1000 simulations (not clustered by session without fixed effects). None of the experimental control variables (confederate delivering the second testimonial, participant gender, and participant age) influenced the number of leader points allocated to the second presenter. We depict these results in Figure 3. Attitudinal ratings of status of the second presenter were significantly higher when he delivered the humorous testimonial (M ⫽ 4.46, SD ⫽ 1.23) than when he delivered the serious testimonial (M ⫽ 4.23, SD ⫽ 1.06), t(15) ⫽ 4.21, p ⬍ .01, 95% CI [0.19, 0.58].3 Competence. Ratings of competence of the second presenter were also significantly higher when the confederate delivered the humorous testimonial (M ⫽ 5.14, SD ⫽ 1.17) than when the confederate delivered the serious testimonial (M ⫽ 4.90, SD ⫽ 1.05), t(15) ⫽ 2.70, p ⬍ .05, 95% CI [0.07, 0.59]. Confidence. We find that the second presenter was rated as more confident when he delivered a humorous testimonial (M ⫽ 5.49, SD ⫽ 1.19) than when he delivered a serious testimonial

(M ⫽ 4.94, SD ⫽ 1.27), t(15) ⫽ 5.67, p ⬍ .001, 95% CI [0.53, 1.17]. For ratings of status, confidence, and competence of the second presenter, we found a significant effect for the confederate who presented (ps ⬍ .05). These effects are driven by one confederate who received low ratings of status, competence, and confidence. Though this confederate received low ratings for each of our dependent variables, he still received higher ratings when he delivered the humorous testimonial than when he delivered the serious testimonial. By controlling for confederate fixed effects, we account for this confederate’s low baseline ratings in our analysis. We also find a significant effect of participant gender on ratings of status of the second presenter, t(15) ⫽ ⫺2.17, p ⬍ .05; men rated the second presenter lower on the attitudinal measure of status. Age of the participant did not influence ratings of status, competence, or confidence of the second presenter. We report results including all of our data. Mediation. In both Baron and Kenny (1986) and bootstrap analyses (Preacher & Hayes, 2004, 2008), we find that competence mediated the relationship between the second presenter’s testimonial and our behavioral and attitudinal measures of status, and that confidence mediated the relationship between the second presenter’s testimonial and our attitudinal measure of status (see Appendix F and Table 5). Although the indirect effect of confidence on status is consistently positive and significant in our other studies (see Table 5), confidence did not mediate the relationship between the second presenter’s testimonial and our behavioral measure of status in this study. Summary. In Study 1b, with both an attitudinal and a behavioral measure of status, we found that individuals show deference to humorous individuals. When an individual makes a comment that is funny and appropriate, others view that individual as more confident and competent and are more likely to select them as a group leader.

Discussion In a face-to-face interaction, the use of humor can increase perceptions of the joke teller’s confidence and competence. By appearing more confident and competent, the joke teller was viewed as higher in status. In Study 1b, we found that by signaling competence, the joke teller was also more likely to be selected for a leadership position in a subsequent task.

Study 2: Joke Success as a Moderator In Study 2, we explore humor in different contexts and we consider a boundary condition that may moderate the influence of humor on status: joke success. The decision to tell a joke may be

Figure 3.

Status conferral in Study 1b: Switzerland Testimonial.

3 When we control for participant ratings of the first presenter, but do not cluster by session or control for confederate fixed effects, the effects are significant for the behavioral measure of status (p ⬍ .05), the attitudinal measure of status (p ⬍ .01), ratings of competence (p ⬍ .001), and ratings of confidence (p ⬍ .001). When we do not control for the first presenter, do not cluster by session, and do not control for confederate fixed effects, the effects remain significant for the behavioral measure of status (p ⬍ .05), are not significant for the attitudinal measure of status (p ⫽ .16), are not significant for competence (p ⫽ .14), and are significant for confidence (p ⬍ .01).

HUMOR AND STATUS

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

risky. In Studies 2a and 2b, we explore perceptions of joke tellers when their joke is appropriate, but fails to elicit laughter. In Study 2a, we consider positive affect as a mediator of the relationship between humor and status. An appropriate, funny joke may induce positive affect in the audience, and positive affect could boost the audience’s evaluation of the joke teller. In Study 2b, we test confidence and competence as mediators of the relationship between humor and status. Telling a joke is likely to make a joke teller appear more confident, but we expect only successful jokes—those that elicit laughter—to cause a joke teller to appear more competent.

9

were appropriate (not offensive), so whether or not the jokes participants recalled failed or succeeded should be related to whether or not participants viewed the jokes as boring. If a humor attempt by the coworker failed by being too benign, it is likely that participants would not have recalled the coworker’s comment as a joke at all. Finally, we asked participants to report the relative rank of their coworker. We asked participants to characterize their coworker’s rank as senior, equal, or subordinate to them in their organization because the relative status of a coworker might impact how funny their jokes seem (e.g., a participant might rate a joke told by a manager as funnier than a joke told by a subordinate).

Study 2a Method Participants. We recruited 120 participants online via Amazon.com’s Mechanical Turk to participate in a short survey in exchange for $0.50 (70% male, Mage ⫽ 31.54 years, SD ⫽ 8.63). Design and procedure. We randomly assigned participants to one of three between-subjects conditions: Successful Joke versus Failed Joke versus Serious Comment. Across all conditions, we asked participants to think of five coworkers they had known for less than a year. Participants wrote down the first name and last initial for each coworker. We then asked all participants to think about the third coworker they wrote down. We asked participants in the Serious Comment condition to recall the last greeting this coworker told them. We asked participants in the Successful Joke condition to recall the last appropriate joke this coworker told them that the participant thought was funny. We gave the participants in the Failed Joke condition nearly identical instructions as the Successful Joke condition. However, in the Failed Joke condition, we asked participants to recall a joke they thought was not funny. We asked participants in all three conditions to write about what coworker 3 had said with enough detail that someone who did not know them or their coworker could understand their coworker’s comments. After writing about what their coworker had said, we asked participants to rate, using a 7-point scale (1 ⫽ not at all, 7 ⫽ extremely), their coworker on the following qualities: respected, admired, and influential. We combined these three items to form the same status index we used in Study 1b (adapted from Anderson et al., 2012; Kilduff & Galinsky, 2013; ␣ ⫽ .90). To measure affect, we asked participants to complete the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988). The PANAS consists of 20 items. Ten items of the PANAS measure positive affect: interested, excited, strong, enthusiastic, proud, alert, inspired, determined, attentive, and active (␣ ⫽ .91). The other 10 items measure negative affect: distressed, upset, guilty, scared, hostile, irritable, ashamed, nervous, jittery, and afraid (␣ ⫽ .94). We asked participants to indicate, on a 7-point scale (1 ⫽ not at all, 7 ⫽ extremely), to what extent they felt that way at the present moment. Next, we asked participants to complete a manipulation check. To measure the funniness of the comments recalled, we asked participants to rate the extent to which coworker 3’s comments were “funny” and “humorous” (r ⫽ .94). We also had participants rate the extent to which coworker 3’s comments were “boring” and “dull” (r ⫽ .87). We instructed participants to recall jokes that

Results and Discussion Manipulation check. Our humor manipulation was successful. Participants rated their coworker’s comment as significantly funnier in the Successful Joke condition than in the Failed Joke condition and the Serious Comment condition. Funniness ratings were also significantly lower in the Failed Joke condition than they were in the Serious Comment condition. Participants rated their coworker’s comment as significantly less boring in the Successful Joke condition than in the Failed Joke condition and the Serious Comment condition. Status. We conducted an analysis of variance (ANOVA) on status ratings as a function of experimental condition: Successful Joke versus Failed Joke versus Serious Comment. We found a significant main effect of experimental condition on ratings of status. Participants’ ratings of their coworker’s status were significantly higher in the Successful Joke condition than they were in the Failed Joke condition and the Serious Comment condition. We depict these results in Figure 4. Other differences in perceived status were not significant. The difference between the Serious Comment and Failed Joke conditions was directional, but not significant (p ⫽ .42). Controlling for the coworker’s relative rank did not significantly alter any of our results. We summarize the results of Study 2a in Table 3. Mediation. We conducted both Baron and Kenny (1986) and bootstrap analyses (Preacher & Hayes, 2004, 2008). The relationship between humor and status was not mediated by affect; that is, positive or negative affect cannot account for the boost in percep-

Figure 4. Joke teller status in Study 2a: Appropriate Joke Recalled.

BITTERLY, BROOKS, AND SCHWEITZER

10

Table 3 Summary of Results for Studies 2a, 2b, 3a, and 3b Joke

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Variable Study 2a Funny Boring Status PA NA Study 2b Status Competence Confidence Study 3a Status Competence Confidence Study 3b Status Competence Confidence

␩2

F

95% CI

Power

Serious response M (SD)

Successful M (SD)

Failed M (SD)

46.24ⴱⴱⴱ 8.17ⴱⴱⴱ 5.42ⴱⴱ 1.99 .31

.44 .12 .08 .03 .01

[.30, [.03, [.01, [.00, [.00,

.54] .23] .18] .11] .04]

1.00 .97 .87 .40 .09

2.62a (1.70) 3.39a (2.04) 4.18a (1.40) 3.65a (1.32) 1.88a (1.31)

5.09b (1.39) 2.43b (1.48) 4.94b (1.18) 3.89ab (1.36) 1.83a (1.16)

1.95c 4.20a 3.94a 4.22b 1.68a

F(2, 271) ⫽ 16.12ⴱⴱⴱ F(2, 271) ⫽ 7.95ⴱⴱⴱ F(2, 271) ⫽ 28.15ⴱⴱⴱ

.11 .06 .17

[.04, .17] [.01, .11] [.09, .25]

1.00 .94 1.00

4.15a (1.08) 4.73a (1.00) 4.73a (1.31)

4.95b (0.83) 5.21b (0.94) 6.10b (1.08)

4.22a (1.19) 4.63a (1.21) 5.52c (1.32)

F(2, 271) ⫽ 37.73ⴱⴱⴱ F(2, 271) ⫽ 59.78ⴱⴱⴱ F(2, 271) ⫽ 3.31ⴱ

.22 .31 .02

[.13, .30] [.22, .38] [.00, .07]

1.00 1.00 .64

4.58a (1.08) 5.08a (0.92) 5.25a (1.36)

3.86b (0.96) 3.92b (1.15) 5.70b (0.96)

3.30c (0.93) 3.27c (1.29) 5.46ab (1.18)

F(2, 225) ⫽ 14.75ⴱⴱⴱ F(2, 225) ⫽ 26.95ⴱⴱⴱ F(2, 225) ⫽ 7.18ⴱⴱ

.12 .19 .06

[.05, .19] [.11, .28] [.01, .12]

1.00 1.00 .94

4.20a (1.00) 4.65a (1.15) 4.64a (1.43)

3.97a (1.29) 3.69b (1.50) 5.52b (1.66)

3.27b (0.98) 3.06c (1.34) 5.37b (1.48)

F(2, F(2, F(2, F(2, F(2,

117) 117) 117) 117) 117)

⫽ ⫽ ⫽ ⫽ ⫽

(1.20) (1.96) (1.46) (1.24) (0.98)

Note. Means in each column with different subscripts are significantly different at the p ⬍ .05 level. We present the simulated power at an ␣ of .05 using 1,000 simulations. ⴱ p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.

tions of status triggered by recalling a successful joke told by a coworker (see Appendix F and Table 5). Summary. In this study, we asked participants to recall an exchange they had with a coworker. We found that coworkers in the humorous conditions recalled a wide array of jokes. The jokes participants recalled significantly influenced their perceptions of their coworker’s status. Recalling an appropriate joke increased perceptions of status in an existing relationship, but only if the joke was successful. By having participants recall jokes, we were able to test the effects of many different joke stimuli and the effects of humor in existing relationships. Our design, however, has limitations. We did not find that positive or negative affect mediate the relationship between humor and status, but participants may not have experienced the same affect during recall that they felt at the time the joke was told. Furthermore, recalling a successful or unsuccessful joke told by a coworker may have increased the salience of positive or negative traits of the coworker. Notably, even transitory shifts in perceptions of status may have lasting effects. For example, if perceptions of status shift during the course of a group decision-making process, the relative influence individuals exert is likely to change, and these changes may influence outcomes. Though documenting the influence of humor in existing relationships is a strength of this study, the possibility of misattribution is a limitation. Though participants did not report having a difficult time recalling a coworker’s joke, it is possible that participants misattributed jokes and recalled a joke told by someone different from the third coworker they listed, and rated that person instead. In our remaining studies, we hold the joke teller constant and manipulate the joke in order to establish a clear causal link between humor attempts and perceptions of status.

Study 2b We extend our investigation of joke success in Study 2b in a different context, using a different joke than we used in Studies 1a and 1b, and by manipulating audience laughter as an indicator of joke success. We also examine confidence and competence as mechanisms of the relationship between humor and status. Although both successful and unsuccessful humor attempts should make a joke teller appear more confident, only successful humor attempts should make a joke teller appear more competent.

Method Participants. We recruited 274 participants online via Amazon.com’s Mechanical Turk to participate in a short survey in exchange for $0.25 (55% male, Mage ⫽ 31.45 years, SD ⫽ 11.25). Design and procedure. We randomly assigned participants to one of three between-subjects conditions: Successful Joke versus Failed Joke versus Serious Response. For our dependent measures, we used the same items for status (␣ ⫽ .81), competence (␣ ⫽ .92), and confidence as we used in Study 1a. In this study, we asked participants to imagine a job candidate interviewing with a manager. The manager asks the candidate a question (“Where do you see yourself in five years?”) and the candidate responds with either a Serious Response (“Continuing to work in this field in a role like this one”) or a Joke we adapted from comedian Mitch Hedberg (“Celebrating the fifth year anniversary of you asking me this question”; quoted in Thinkexist.com, 2014). We manipulated the success of the joke by describing the manager’s response. After the joke, participants were informed that the manager either laughed or sat in silence. We include an example screenshot from this scenario in Appendix E.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

HUMOR AND STATUS

11

Results and Discussion

Discussion

We identify audience laughter as a key moderator of the relationship between humor and status. We find that appropriate humor attempts increase status as long as they are successful (i.e., the manager laughs). We report the means, standard deviations, and test statistics of Study 2b in Table 3. Status. Participants’ ratings of the interviewee’s status were significantly higher in the Successful Joke condition than in the Failed Joke condition and the Serious Response condition (ps ⬍ .0001). The difference in perceived status between the Serious Response and Failed Joke conditions was not significant (p ⫽ .66). Competence. Participants’ ratings of the interviewee’s competence were significantly higher in the Successful Joke condition than in the other two conditions (ps ⬍ .01). We depict these results in Figure 5. Confidence. Participants’ ratings of the interviewee’s confidence were highest in the Successful Joke condition and lowest in the Serious Response condition. We report confidence ratings across conditions in Table 3. In planned pairwise comparisons, ratings of the interviewee’s confidence were significantly different across all conditions (ps ⬍ .01). Mediation. We conducted both Baron and Kenny (1986) and bootstrap analyses (Preacher & Hayes, 2004, 2008) and found that both perceptions of confidence and competence mediated the relationship between the successful use of humor and status (see Appendix F and Table 5). However, whereas confidence was significantly higher in both joke conditions than the Serious Response condition, competence and status were only higher in the Successful Joke condition. We find that the indirect effect of confidence was positive and significant regardless of whether or not the joke was successful. The indirect effect of competence, however, was only positive and significant if the joke is successful. Summary. In this study, we again identify perceptions of confidence and competence as the mechanisms linking the successful use of humor and status. Attempting to use humor made the joke teller appear more confident, whether or not the joke was successful. However, only an appropriate, successful joke increased perceived competence and boosted status.

In Study 2, we extended our investigation of humor and status with different methods. In Studies 2a and 2b, we identify joke success (i.e., audience laughter) as an important moderator of the relationship between humor and competence. A humor attempt does not enhance perceptions of competence and status when the audience does not find it funny. Interestingly, in our studies, when the audience did not find the joke funny, the humor attempt did not harm status compared with the no-humor-attempt condition. We speculate that the “failed” jokes in this study were not large failures, because they were generally funny and appropriate. This was certainly true of the humor attempt in Study 2b (see joke ratings from Pilot Study 2, summarized in Table 1).

Figure 5.

Study 3: Inappropriate Jokes as a Boundary Condition In Studies 3a and 3b, we extend our investigation to the use of inappropriate jokes. As in Study 2b, we manipulate the success of humor attempts by describing an audience who either laughs or does not laugh. Across both Studies 3a and 3b, we present participants with jokes that were judged by participants in Pilot Study 2 to be inappropriate for an interview. We consider the prospect that telling an inappropriate, unsuccessful joke demonstrates confidence but signals a lack of competence and may actually decrease status.

Study 3a Method Participants. We recruited 274 participants online via Amazon.com’s Mechanical Turk to participate in a short survey in exchange for $0.25 (57% male, Mage ⫽ 30.03 years, SD ⫽ 9.94). Design and procedure. The design of Study 3a was nearly identical to Study 2, except for the manager’s final question and the candidate’s response. We randomly assigned participants to one of three between-subjects conditions: Successful Joke versus Failed Joke versus Serious Response.

Joke teller status, competence, and confidence in Study 2b: Five-Year Anniversary Joke.

BITTERLY, BROOKS, AND SCHWEITZER

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

12

Across all three conditions, the manager asked the candidate, “Are you looking for a challenging position?” In the Serious Response condition, the candidate responded by saying, “Yes. I am a hard worker and like challenges.” In the Successful Joke condition, the job candidate answered the manager’s question with a joke rated as inappropriate in Pilot Study 2. Specifically, the candidate replied by saying, “That’s what she said!” and participants then read that “The manager and candidate both laugh.” The Failed Joke condition used the same candidate response, “That’s what she said!” but this time “The candidate laughs and the manager sits in silence.” In this way, both joke conditions used an inappropriate joke, but we manipulated the success of the joke by changing the manager’s reaction (laughter vs. no laughter). In both joke conditions, the candidate then adds, “But seriously, yes. I am a hard worker and like challenges.” We used the same items for status (␣ ⫽ .77), competence (␣ ⫽ .94), and confidence as we used in Studies 1a and 2b.

the Serious Response condition. Ratings of the interviewee’s confidence in the Failed Joke condition were not significantly different from those in the Serious Response condition or the Successful Joke condition. We depict these results in Figure 6. Mediation. We conducted both Baron and Kenny (1986) and bootstrap analyses (Preacher & Hayes, 2004, 2008) and found that perceptions of competence mediated the relationship between the use of inappropriate humor and status. Confidence mediated the relationship between the successful use of inappropriate humor and status (see Appendix F and Table 5). Summary. In this study, we identify inappropriate humor as a boundary condition of the positive relationship between humor and status. Compared to not using humor, making an inappropriate joke caused the job candidate to be viewed as more confident. However, making an inappropriate joke caused the job candidate to appear less competent, which in turn lowered status. This effect was even more dramatic when the candidate made an inappropriate joke and the manager did not laugh.

Results and Discussion We find that inappropriate humor attempts make a joke teller appear more confident, but less competent and decrease status. However, audience laughter reduces the harmful effects of telling an inappropriate joke. We report the means, standard deviations, and test statistics for Study 3 in Table 3. Status. Participants’ ratings of the interviewee’s status were highest in the Serious Response condition, lower in the Successful Joke condition, and lowest in the Failed Joke condition. In planned pairwise comparisons, ratings of status were significantly different across all three conditions (ps ⬍ .001). Competence. Participants’ ratings of the interviewee’s competence were highest in the Serious Response condition, lower in the Successful Joke condition, and lowest in the Failed Joke condition. In planned pairwise comparisons, competence levels were significantly different across each of the three conditions (ps ⬍ .001). Confidence. We found a different pattern of results looking at confidence. Participants’ ratings of the interviewee’s confidence were significantly higher in the Successful Joke condition than in

Figure 6. priate).

Study 3b We conducted a conceptual replication of Study 3a with a different inappropriate joke and a different participant pool. We recruited 228 adults from a city in the northeastern United States to participate in a study in exchange for $10 in a behavioral laboratory (42% male, Mage ⫽ 23.79 years, SD ⫽ 9.40). In all conditions, the manager asked the candidate, “What do you see yourself doing in the first 30 days of this job?” In the Serious Response condition, the candidate responded by saying, “Getting to know the team and getting up to speed.” In the humor conditions, the candidate replied by saying, “The receptionist I saw on the way in.”

Results As we found in Study 3a, results from Study 3b demonstrate that telling an inappropriate joke can decrease status, compared to not making a joke. We report the results for this study in Table 3, and

Joke teller status, competence, and confidence in Study 3a: That’s What She Said Joke (inappro-

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

HUMOR AND STATUS

Figure 7.

13

Joke teller status, competence, and confidence in Study 3b: Receptionist Joke (inappropriate).

we depict these results in Figure 7. We find that telling a joke signals confidence, which typically boosts perceptions of status. Telling an inappropriate joke, however, signals low competence in addition to high confidence. In our studies, the signal of low competence outweighed the signal of confidence, and participants judged targets who told inappropriate jokes to have lower status. That is, the combined effects of confidence and low competence decreased status. These results provide further support for the importance of the manager’s reaction; telling a joke that elicits laughter signals a greater level of competence than telling a joke that elicits no laughter. The manager’s laughter mitigates the harmful effect of telling an inappropriate joke on perceptions of the job candidate’s status. When the candidate told an inappropriate joke, the candidate was seen as more competent and higher status when the manager laughed than when the manager did not laugh.

Discussion Findings from Study 3 support our conceptualization of humor as risky. Merely attempting to use humor makes an individual appear confident, but the appropriateness and success of the attempt influence perceptions of the joke teller’s competence. Whereas appropriate jokes signal competence and boost status (Studies 1–2), inappropriate jokes signal low competence and can decrease status (Study 3). Eliciting laughter with an inappropriate joke mitigates the harmful effects of telling an inappropriate joke on status.

Study 4: Comparing Appropriate and Inappropriate Humor Attempts In Study 4, we investigate the effect of joke success and the effect of joke appropriateness simultaneously. In Study 4a, we examine confidence and competence as the mechanisms linking the use of humor with changes in status. In Study 4b, we rule out positive and negative affect as an alternative explanation for the relationship between humor and status. If an inappropriate joke induces negative affect in the audience, the negative affect might

harm ratings of a joke teller’s competence and status. In Study 4b, we also disentangle the effects of funniness and appropriateness. We contrast the consequences of jokes that are similarly funny, but very different with respect to appropriateness (see results from Pilot Study 2, summarized in Table 1). We test whether or not joke appropriateness moderates the relationship between humor and status.

Study 4a Method Participants. We recruited 186 adults from a city in the northeastern United States to participate in a study in exchange for $10 in a behavioral lab (34% male, Mage ⫽ 20.10 years, SD ⫽ 2.10). Design and procedure. In Study 4a, we randomly assigned participants to one of five between-subjects conditions: Appropriate Successful Joke versus Appropriate Failed Joke versus Inappropriate Successful Joke versus Inappropriate Failed Joke versus Serious Response. Across all five conditions, the manager asked the candidate, “What would you do if you won the lottery?” In the Serious Response condition, the candidate responded by saying, “I would probably go on a vacation to Hawaii.” In the Appropriate Joke conditions, the job candidate answered the manager’s question by saying, “When I die, I would want my last words to be, ‘I left one million dollars under the. . .’” In the Inappropriate Joke conditions, the candidate answered the manager’s question with, “I’ll tell you what I’d do, two chicks at the same time” (quoted in IMDb.com, 2015). In the Successful Joke conditions, after the joke, the participants read that, “The manager and candidate both laugh.” In the Failed Joke conditions, the participants are informed that, “The candidate laughs and the manager sits in silence.” In all four joke conditions, the scenario ends with the candidate saying, “But seriously, I would probably go on a vacation to Hawaii.” For our dependent variables, we used the same status (␣ ⫽ .69), the competence (␣ ⫽ .92), and confidence items that we used in our prior

BITTERLY, BROOKS, AND SCHWEITZER

14

studies. We also asked participants to rate, on a 7-point scale (1 ⫽ not at all, 7 ⫽ extremely), the candidate on nine dimensions. Seven of the items were filler items. The two items of interest were “funny” and “inappropriate.” The other seven items (agreeable, interesting, thoughtful, persuasive, dominant, pleasant, and considerate) were used to mask the purpose of the study and were not analyzed.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Results and Discussion We find that a successful, appropriate humor attempt makes a joke teller appear more competent and increases status, but a failed, inappropriate humor attempt causes a joke teller to appear less competent and harms status. We find that all humor attempts cause the joke teller to appear more confident, which helps status. We find an effect of laughter; joke tellers are perceived to be more confident and competent when the audience laughs than when the audience does not laugh. We report the means, standard deviations, and test statistics of Study 4a in Table 4. Manipulation checks. Participants rated successful jokes as funnier than unsuccessful jokes. Participants rated the inappropriate joke as far more inappropriate than the appropriate joke. The serious response was rated as the most appropriate response. Status. The successful, appropriate joke increased ratings of the candidate’s status, but the failed, inappropriate joke decreased status (see Figure 8). Ratings of the interviewee’s status were significantly higher in the Appropriate Successful Joke condition than in all other conditions (ps ⬍ .001). Ratings of status were also significantly lower in the Inappropriate Failed Joke condition than in all other conditions (ps ⬍ .001). Competence. Similar to the results for status, participants rated the candidate’s competence highest after a successful, appropriate joke and lowest after a failed, inappropriate joke. Ratings of the interviewee’s competence were significantly higher in the Appropriate Successful Joke condition than in all other conditions (ps ⬍ .001). Ratings of competence were also significantly lower in the Inappropriate Failed Joke condition than in all other conditions (ps ⬍ .01). We depict this pattern of results in Figure 8.

Confidence. Confidence ratings were higher in all four joke conditions than they were in the Serious Response condition (ps ⬍ .05). Mediation. Perceptions of confidence and competence fully mediated the relationship between the Appropriate Successful Joke condition and status. Regardless of joke outcome, the indirect effect of confidence is positive and significant (see Appendix F and Table 5). However, joke appropriateness and success moderate the indirect effect of competence. The indirect effect of competence was negative, but not significant, after an appropriate joke that fails. The indirect effect of competence was negative and significant for an inappropriate joke, regardless of outcome. These results are consistent with our model (see Figure 1). Summary. In this study, we found that telling an appropriate joke that elicits laughter increased status, but telling an inappropriate joke that fails to elicit laughter harmed status. We found that confidence and competence mediate the relationship between the successful use of appropriate humor and status. However, the appropriateness and success of a joke changes perceptions of competence. Individuals who tell both appropriate and inappropriate jokes are perceived to be more confident than those who tell no jokes, but only individuals who tell appropriate jokes that elicit laughter are perceived to be more competent than those who tell inappropriate jokes that fail to elicit laughter and those who tell no jokes at all.

Study 4b In Study 4b, we extend our investigation of inappropriate jokes. In this study, we use different jokes, all of which were rated as very funny, but differ with respect to appropriateness. In this study, we use the same status measure as the one we used in Studies 1b and 2a, and we test whether joke appropriateness moderates the relationship between humor and status. We also consider whether affect mediates the relationship between humor attempts and status

Table 4 Summary of Results for Studies 4a and 4b Appropriate Joke Variable Study 4a Funny Inappropriate Status Competence Confidence Study 4b Funny Boring Appropriate Status PA NA

␩2

F

95% CI

Inappropriate Joke

Power

Serious response M (SD)

Successful M (SD)

Failed M (SD)

Successful M (SD)

Failed M (SD)

F(4, F(4, F(4, F(4, F(4,

181) 181) 181) 181) 181)

⫽ ⫽ ⫽ ⫽ ⫽

17.64ⴱⴱⴱ 80.14ⴱⴱⴱ 17.36ⴱⴱⴱ 26.34ⴱⴱⴱ 9.33ⴱⴱⴱ

.28 .64 .28 .37 .17

[.16, [.55, [.16, [.25, [.07,

.37] .69] .36] .45] .25]

1.00 1.00 1.00 1.00 1.00

3.29ac (1.47) 2.13a (1.36) 4.07a (0.84) 3.95a (0.98) 4.29a (1.29)

5.65b (1.09) 2.97b (1.24) 4.62b (0.88) 4.45b (0.98) 5.78bc (1.11)

3.89a (1.39) 4.16c (1.66) 3.68a (0.76) 3.74a (0.84) 5.03d (1.46)

3.57ac (1.85) 6.35d (1.11) 3.73a (0.93) 3.00c (1.10) 6.00b (1.22)

2.92c (1.77) 6.35d (1.14) 2.97c (0.99) 2.34d (0.99) 5.30cd (1.61)

F(4, F(4, F(4, F(4, F(4, F(4,

504) 504) 504) 504) 504) 504)

⫽ ⫽ ⫽ ⫽ ⫽ ⫽

47.94ⴱⴱⴱ 38.10ⴱⴱⴱ 132.32ⴱⴱⴱ 45.23ⴱⴱⴱ .77 1.82

.28 .23 .51 .26 .01 .01

[.21, [.17, [.45, [.20, [.00, [.00,

.33] .29] .56] .32] .02] .03]

1.00 1.00 1.00 1.00 0.25 0.59

2.16a (1.40) 3.74a (1.66) 5.96a (1.01) 3.94a (1.27) 4.03a (1.23) 1.60ab (0.98)

5.07b 1.82b 5.51b 4.84b 4.03a 1.38b

4.28c (1.61) 2.24c (1.33) 4.43c (1.20) 3.27c (1.27) 4.06a (1.29) 1.68a (0.99)

4.16c (1.67) 1.92bc (1.12) 3.04d (1.65) 3.61ac (1.50) 3.93a (1.47) 1.65a (0.92)

4.25c (1.80) 1.92bc (1.20) 2.64e (1.47) 2.57d (1.33) 3.79a (1.32) 1.63a (0.88)

(1.39) (1.16) (1.00) (0.95) (1.33) (0.68)

Note. Means in each column with different subscripts are significantly different at the p ⬍ .05 level. We present the simulated power at an ␣ of .05 using 1,000 simulations. ⴱ p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.

HUMOR AND STATUS

15

Table 5 Summary of Mediation Results Study

Condition

Ind. effect

1a 1b

Appropriate successful joke Leadership points Attitudinal status Appropriate successful joke Appropriate failed joke Inappropriate successful joke Inappropriate failed joke Inappropriate successful joke Inappropriate failed joke Appropriate successful joke Appropriate failed joke Inappropriate successful joke Inappropriate failed joke

.32 ⫺.28 .14 .28 .16 .09 .04 .17 .14 .30 .15 .35 .20

Appropriate successful joke Appropriate failed joke Appropriate successful joke Appropriate failed joke Inappropriate successful joke Inappropriate failed joke

.09 .20 .00 .01 ⫺.03 ⫺.08

95% CI

Ind. effect

Confidence

2b 3a

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

3b 4a

[.18, .50] [⫺1.07, .33] [.04, .27] [.16, .44] [.07, .29] [.03, .17] [⫺.03, .13] [.07, .31] [.05, .27] [.17, .48] [.02, .31] [.19, .56] [.07, .40]

Positive affect 2a 4b

[⫺.11, .35] [.03, .51] [⫺.11, .11] [⫺.10, .12] [⫺.15, .09] [⫺.20, .03]

95% CI

Ind. effect

Competence .11 .37 .16 .29 ⫺.06 ⫺.67 ⫺1.05 ⫺.47 ⫺.79 .27 ⫺.11 ⫺.52 ⫺.88

Total

[.03, .24] [.03, 1.06] [.05, .31] [.12, .48] [⫺.26, .14] [⫺.87, ⫺.50] [⫺1.29, ⫺.84] [⫺.73, ⫺.26] [⫺1.05, ⫺.56] [.03, .54] [⫺.34, .11] [⫺.80, ⫺.26] [⫺1.20, ⫺.59]

.43 .09 .30 .58 .11 ⫺.58 ⫺1.01 ⫺.30 ⫺.64 .57 .03 ⫺.17 ⫺.67

Negative affect .00 .00 ⫺.01 .00 .00 .00

[⫺.05, [⫺.03, [⫺.06, [⫺.01, [⫺.01, [⫺.01,

95% CI

[.27, .62] [⫺.62, .82] [.14, .48] [.33, .82] [⫺.17, .38] [⫺.81, ⫺.36] [⫺1.28, ⫺.76] [⫺.62, .00] [⫺.96, ⫺.34] [.27, .90] [⫺.25, .32] [⫺.54, .20] [⫺1.08, ⫺.29] Total

.08] .11] .01] .04] .04] .03]

.09 .21 ⫺.01 .01 ⫺.03 ⫺.08

[⫺.12, .35] [.03, .53] [⫺.13, .10] [⫺.09, .13] [⫺.15, .09] [⫺.20, .03]

Note. Comparisons in each row reflect contrasts with the Serious condition. We report the indirect effects using 5,000 simulation bootstrap analysis (Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008).

by exploring if the harm to a joke teller after an inappropriate joke might be driven by negative affect felt by observers.

Method Participants. We recruited 509 adults from Amazon.com’s Mechanical Turk to participate in a study in exchange for $0.45 (52% male, Mage ⫽ 33.89 years, SD ⫽ 11.25). Design and procedure. In Study 4b, we randomly assigned participants to one of five between-subjects conditions: Appropriate Successful Joke versus Appropriate Failed Joke versus Inappropriate Successful Joke versus Inappropriate Failed Joke versus Serious Response. Across all five conditions, the manager asked the candidate, “What is a creative use for an old tire?” In the Serious Response condition, the candidate responded by saying, “Make a tire swing out of it.” In the Appropriate Joke conditions, the job candidate answered the manager’s question by saying, “Someone doing CrossFit could use it for 30 minutes, then tell you about it forever.” In the Inappropriate Joke conditions, the candidate answered the manager’s question with, “Melt it down, make 365 condoms, and call it a GOODYEAR!” In the Successful Joke conditions, after the joke, the participants read that, “The manager and candidate both laugh.” In the Failed Joke conditions, the participants read, “The candidate laughs and the manager sits in silence.” In all four joke conditions, the scenario ends with the candidate saying, “But seriously, make a tire swing out of it.” For our dependent variables, we used the same status index (adapted from Anderson et al., 2012; Kilduff & Galinsky, 2013; ␣ ⫽ .95) as the one we used in Studies 1b and 2a. After rating status, participants completed the PANAS (Watson, Clark, & Tellegen, 1988; ␣PA ⫽ 0.93; ␣NA ⫽ 0.94). We also asked partic-

ipants to rate, on a 7-point scale (1 ⫽ not at all, 7 ⫽ extremely), the candidate’s response to the manager’s question on eight dimensions (funny, humorous, boring, dull, inappropriate, appropriate, tasteless, and suitable). We combined two items (funny, humorous) to create a measure of funniness (r ⫽ .90) and another two items (boring, dull) to measure boringness (r ⫽ .83). We combined the remaining items (inappropriate (reverse scored), appropriate, tasteless (reverse scored), suitable) to form a measure of appropriateness (␣ ⫽ .93).

Results and Discussion As in Study 4a, we find that joke appropriateness moderates the relationship between humor and status. With different stimuli, we find that a successful, appropriate humor attempt increases the joke teller’s status, but a failed, inappropriate humor attempt harms the joke teller’s status. Although affect could cause participants to rate the joke teller more or less favorably, we do not find that affect mediates the relationship between humor and status. We report the means, standard deviations, and test statistics of this study in Table 4. Manipulation checks. The candidate’s response was rated as significantly funnier and less boring in all of the joke conditions compared to the Serious Response condition (all ps ⬍ .0001). Participants judged the candidate’s response to be significantly more appropriate in the Appropriate Joke conditions (Mappropriateness ⬎ 4.42 for the Appropriate Successful and Appropriate Failed conditions) than in the Inappropriate Joke conditions (Mappropriateness ⬍ 3.05 for the Inappropriate Successful and Inappropriate Failed conditions; ps ⬍ .0001). Participants judged successful humor attempts to be more appropriate than failed humor attempts.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

16

BITTERLY, BROOKS, AND SCHWEITZER

Figure 8. The benefits and risk of humor. (A) Joke teller status in Study 4a. (B) Joke teller competence in Study 4a. (C) Joke teller confidence in Study 4a.

Status. Ratings of the interviewee’s status were significantly higher in the Appropriate Successful Joke condition than in all other conditions (ps ⬍ .001). Ratings of status were also significantly lower in the Inappropriate Failed Joke condition than in all other conditions (ps ⬍ .001). We depict these results in Figure 9. Mediation. We conducted both Baron and Kenny (1986) and bootstrap analyses (Preacher & Hayes, 2004, 2008) and we did not find that affect mediated the relationship between humor and status (see Appendix F and Table 5). Positive affect did not boost ratings of status after a successful, appropriate joke, and negative affect did not diminish status after a failed, inappropriate joke. Moderation. We tested the moderating effect of appropriateness on the relationship between humor and status. We conducted an analysis of variance (ANOVA) on status ratings as a function of experimental condition, appropriateness, and the interaction of experimental condition and appropriateness, F(9, 499) ⫽ 45.34, p ⬍ .0001, ␩2 ⫽ .45, 95% CI [0.38, 0.49]. The effect of experimental condition on ratings of status was marginally significant, F(4, 499) ⫽ 2.38, p ⫽ .05, ␩2 ⫽ .02, the effect of ratings of appropriateness of the response was significant, F(1, 499) ⫽ 115.00, p ⬍ .0001, ␩2 ⫽ .19, and the interaction of response and

appropriateness was significant, F(4, 499) ⫽ 2.40, p ⬍ .05, ␩2 ⫽ .02. Summary. In Study 4b, we extend our investigation of inappropriate humor with jokes that were rated as very funny in Pilot Test 2 (Mfunny ⬎ 4.9 for both jokes). The appropriate joke (CrossFit joke) was rated as appropriate (Mappropriate ⫽ 4.86), but the Goodyear joke was rated as inappropriate (Mappropriate ⫽ 2.62). As in Study 4a, we find that appropriateness moderates the relationship between the use of humor and status. In this study, we also consider and rule out affect as an alternative mechanism; positive affect did not increase status when appropriate jokes elicited laughter, and negative affect did not decrease status when inappropriate jokes failed to elicit laughter. These results are consistent with the findings in Study 2a. In Study 2a, it might have been possible that positive or negative affect did not mediate the relationship between humor and status because participants were not experiencing the same emotions they felt during recall as they felt when the joke was told. In Study 4b, however, we address this concern by measuring positive and negative affect immediately after the humor attempt was delivered.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

HUMOR AND STATUS

Figure 9.

Joke teller status in Study 4b.

Our findings in Studies 4a and 4b illustrate the inherent risk of using humor. Telling an appropriate joke that elicits laughter increases status, but telling an inappropriate joke that fails to elicit laughter decreases status. Telling a joke displays confidence and helps status, but a signal of low competence (e.g., an inappropriate joke) can harm status. These studies also underscore the importance of the audience’s reaction to the joke. When the audience laughs, people are far more likely to perceive the humor attempt as funny and appropriate. These results reveal just how malleable our perceptions of humor are; merely reading that another individual either laughed or did not laugh influences how we evaluate both the humor attempt and the expresser himself.

General Discussion Our findings reveal an important link between humor and status. In Studies 1a and 1b, telling a successful joke— one rated as funny and appropriate—increased the joke teller’s status. Successful joke tellers are viewed as higher in confidence, competence, and status, and are more likely to be nominated as group leaders. Importantly, joke success (i.e., whether or not the audience laughs) moderates the relationship between humor and status. In Study 2a, we found that recalling an appropriate joke told by a coworker increased perceptions of the coworker’s status, but only if the joke was funny. Interestingly, we found the link between humor and status to be so powerful that merely prompting individuals to recall a humorous exchange with a coworker shifted their perceptions of their coworker’s status. That is, in Study 2a we found that merely recalling a humorous exchange shifted perceptions of status in existing relationships. In Study 2b, we used a different experimental paradigm and showed that attempting to use humor displays confidence, but only the successful use of humor signals competence and increases status. In Studies 3a and 3b, we found that the use of humor is risky. Telling an inappropriate joke signals a lack of competence and can decrease status. Even inappropriate jokes, however, signal a high level of confidence. We extended our investigation in Studies 4a and 4b, and found that humor attempts have substantially different effects on status and competence depending on whether or not the joke is appropriate and whether or not the joke elicits laughter. Once again, we found that confidence and

17

competence, not affect, mediate the relationship between humor and status. Taken together, our results demonstrate that humor attempts, even unsuccessful ones, boost perceptions of confidence, but only humor attempts that are appropriate and elicit laughter boost perceptions of competence and status. Inappropriate humor attempts that fail to elicit laughter can overpower the beneficial effects of signaling high confidence and cause a joke teller to appear less competent and harm status. Though humor can boost status, using humor is risky. Humor attempts can fail in several ways: by being too boring (i.e., not funny), too bold (i.e., inappropriate), or failing to elicit laughter from the audience. How the audience reacts profoundly influences perceptions. If the audience does not laugh, observers are less likely to view the humor attempt as appropriate or funny, and the joke teller may lose status.

Theoretical Implications Our findings make several important theoretical contributions. First, we establish an important link between humor and status. Individuals expend substantial resources to gain status. The use of humor, however, may offer a relatively inexpensive, though risky strategy for gaining status by boosting perceptions of confidence and competence. Importantly, our research demonstrates that to understand status, we need to understand humor. Second, our findings describe an important relationship between humor, confidence, and competence. Prior work has focused on how displays of ability, dominance, and confidence signal competence and consequently increase status (Anderson & Kilduff, 2009b; Kennedy, Anderson, & Moore, 2013). We find that merely telling a joke displays confidence, and that perceptions of confidence are associated with higher status. This is consistent with prior work, which has found that displaying confidence can boost status (Anderson et al., 2012; Kennedy et al., 2013). However, we identify the inappropriate and failed use of humor as an important exception. Inappropriate and failed humor attempts display confidence, but simultaneously signal low competence and lower status. That is, failed humor attempts can boost perceptions of confidence, but signal low competence and harm status. Third, our findings underscore the risk of attempting to use humor. Whereas prior humor research has focused on humor attempts that caused other individuals to laugh, we investigate the impact of humor attempts that fail to elicit laughter. Our findings highlight the important role that laughter plays in determining not only whether or not humor attempts succeed, but also how appropriate the use of humor is. Even for objectively inappropriate humor attempts, laughter substantively mitigated the damage that telling an inappropriate joke caused. In general, telling an inappropriate joke signals a lack of competence and damages status. But someone skilled in the ability to elicit laughter may face far fewer consequences for telling inappropriate jokes.

Prescriptive Advice Our results reveal that the ability to use humor is an important social and managerial skill. By using humor effectively, individuals can project confidence, signal competence, and increase their

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

18

BITTERLY, BROOKS, AND SCHWEITZER

status. As a result, individuals within organizations may derive substantial benefits by developing their ability to use humor. Perhaps humor should play an important role in how we select, train, and promote individuals. Our findings also reveal that humor is risky. Using humor to project confidence, signal competence, and increase status may be particularly effective in novel situations when individuals form initial impressions. These settings, however, are also characterized by unfamiliarity. Expressers may fail to appreciate implicit norms and boundaries as they interact with unfamiliar others. It is possible that the contexts in which humor may be most beneficial are also those in which humor is fraught with risk. Ultimately, our prescriptive advice is to use humor with caution.

Future Directions Future work can extend our investigation in several ways. Future research should identify characteristics that moderate the risk of telling an inappropriate joke. To succeed, a joke needs to be both benign (inoffensive) and a violation (surprising/inappropriate enough to make people laugh, McGraw & Warren, 2010; Warren & McGraw, 2016, 2015). Future work should identify guidelines to minimize the risk of telling offensive jokes. For example, aspects of joke delivery (e.g., physical cues, timing, frequency), characteristics of the joke teller (e.g., age, gender, status), the audience (e.g., size, heterogeneity), the target (e.g., present vs. absent, known vs. stranger), the setting (e.g., in the workplace, at home), and the relationships between the joke teller, audience, and target (e.g., hierarchy, length of relationship, social closeness, liking) are all likely to influence how beneficial and risky the use of humor is. Misjudging the context could spell the difference between success and disaster. We found that humor can boost perceptions of confidence, competence, and status. We expect successful joke tellers to be more influential than others. Those who attempt to use humor and fail, however, may lose respect, status, and influence. We call for future work to explore the relationship between successful humor, unsuccessful humor, and influence. Future research should also explore other potential mediators of the relationship between humor and status. In addition to confidence and competence, being able to anticipate what another individual would view as appropriate and humorous reflects social skill. Inferences about social skills may also help to explain why the audience laughing helps to mitigate the negative effects of telling an inappropriate joke. It is also possible that individuals who tell successful, appropriate jokes are better liked than individuals who are serious, whereas those who tell failed, inappropriate jokes are less well-liked. Ultimately, humorous individuals may gain greater influence over time that fuels an even steeper rise in status. In our studies, we operationalized humor attempts with short, witty, spoken jokes. This is a common form of humor expression, but some humor attempts involve other forms of expression such as physical humor or storytelling. Future work should explore how cultural norms and types of humor expression moderate the relationship between humor and status. In our studies, we focused on individual-level outcomes for the joke teller: perceptions of confidence, competence, and status. But humor is likely to impact important outcomes at the dyadic, group,

and organizational levels as well. For example, organizations that encourage the use of humor may be more effective with respect to recruitment and retention than serious organizations. Future work could investigate outcomes at different levels of analysis. Future work should also investigate the moderating role of gender in the relationship between humor and status. Varying the gender of the joke teller, target, and audience may matter profoundly for joke success— especially for gender-related jokes (Feingold, 1992; Hooper, Sharpe, & Roberts, 2016; Martin, 2007; Mickes, Walker, Parris, Mankoff, & Christenfeld, 2012). Future work should explore gender differences across the three humor roles (i.e., joke teller, joke target(s), and joke audience), and how gender differences impact the appropriateness and willingness of individuals to attempt to use humor. Important work remains to guide individuals and groups in how to recover following an inappropriate joke failure. When an individual tells a joke that is inappropriate and unsuccessful, perhaps an apology is the most effective way to regain status. Alternatively, the joke teller’s best recovery strategy might be to make a self-deprecating joke or simply shift focus. In some cases, if the joke is extremely inappropriate (e.g., Justine Sacco’s joke about AIDS in South Africa), the joke teller might not be able to repair the damage done by the joke.

Conclusion Humor is pervasive, and making a joke presents an opportunity for individuals to increase their status. If individuals tell appropriate jokes that make others laugh, they are likely to signal both confidence and competence and increase their status. If individuals tell inappropriate jokes that do not make others laugh, they are likely to appear confident, but less competent and lower in status. Taken together, many individuals may be missing opportunities to project confidence, demonstrate their competence, and increase their status. On the other hand, some individuals may be keenly aware about the risks of making inappropriate jokes— especially at work—and they may be wise to keep their jokes to themselves. Whereas Dick Costolo told jokes as he rose to the top, it only took one inappropriate joke for Justine Sacco to get fired. Humor attempts are risky business.

References Adler, N. E., Epel, E. S., Castellazzo, G., & Ickovics, J. R. (2000). Relationship of subjective and objective social status with psychological and physiological functioning: Preliminary data in healthy white women. Health Psychology, 19, 586 –592. http://dx.doi.org/10.1037/02786133.19.6.586 Anderson, C., Brion, S., Moore, D. A., & Kennedy, J. A. (2012). A statusenhancement account of overconfidence. Journal of Personality and Social Psychology, 103, 718 –735. http://dx.doi.org/10.1037/a0029395 Anderson, C., Hildreth, J. A. D., & Howland, L. (2015). Is the desire for status a fundamental human motive? A review of the empirical literature. Psychological Bulletin, 141, 574 – 601. http://dx.doi.org/10.1037/a0038781 Anderson, C., John, O. P., Keltner, D., & Kring, A. M. (2001). Who attains social status? Effects of personality and physical attractiveness in social groups. Journal of Personality and Social Psychology, 81, 116 –132. http://dx.doi.org/10.1037/0022-3514.81.1.116 Anderson, C., & Kilduff, G. J. (2009a). The pursuit of status in social groups. Current Directions in Psychological Science, 18, 295–298. http://dx.doi.org/10.1111/j.1467-8721.2009.01655.x Anderson, C., & Kilduff, G. J. (2009b). Why do dominant personalities attain influence in face-to-face groups? The competence-signaling ef-

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

HUMOR AND STATUS fects of trait dominance. Journal of Personality and Social Psychology, 96, 491–503. http://dx.doi.org/10.1037/a0014201 Anderson, C., Kraus, M. W., Galinsky, A. D., & Keltner, D. (2012). The local-ladder effect: Social status and subjective well-being. Psychological Science, 23, 764 –771. http://dx.doi.org/10.1177/0956797611434537 Anderson, C., Srivastava, S., Beer, J. S., Spataro, S. E., & Chatman, J. A. (2006). Knowing your place: Self-perceptions of status in face-to-face groups. Journal of Personality and Social Psychology, 91, 1094 –1110. http://dx.doi.org/10.1037/0022-3514.91.6.1094 Anderson, C., Willer, R., Kilduff, G. J., & Brown, C. E. (2012). The origins of deference: When do people prefer lower status? Journal of Personality and Social Psychology, 102, 1077–1088. http://dx.doi.org/10.1037/ a0027409 Avolio, B. J., Howell, J. M., & Sosik, J. J. (1999). A funny thing happened on the way to the bottom line: Humor as a moderator of leadership style effects. Academy of Management Journal, 42, 219 –227. http://dx.doi .org/10.2307/257094 Barkow, J. H., Akiwowo, A. A., Barua, T. K., Chance, M. R. A., Chapple, E. D., Chattopadhyay, G. P., . . . Sarles, H. B. (1975). Prestige and culture: A biosocial interpretation. Current Anthropology, 16, 553–572. http://dx.doi.org/10.1086/201619 Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182. http://dx.doi.org/10.1037/0022-3514.51.6.1173 Berger, J., Cohen, B. P., & Zelditch, M., Jr. (1972). Status characteristics and social interaction. American Sociological Review, 37, 241–255. http://dx.doi.org/10.2307/2093465 Bilton, N. (2012, Oct 6). A master of improv, writing Twitter’s script. The New York Times. Retrieved from http://www.nytimes.com/2012/10/07/ technology/dick-costolo-of-twitter-an-improv-master-writing-its-script .html?pagewanted⫽all&_r⫽0 Brescoll, V. L., & Uhlmann, E. L. (2008). Can an angry woman get ahead? Status conferral, gender, and expression of emotion in the workplace. Psychological Science, 19, 268 –275. http://dx.doi.org/10.1111/j.14679280.2008.02079.x Campos, B., Keltner, D., Beck, J. M., Gonzaga, G. C., & John, O. P. (2007). Culture and teasing: The relational benefits of reduced desire for positive self-differentiation. Personality and Social Psychology Bulletin, 33, 3–16. http://dx.doi.org/10.1177/0146167206293788 Chen, Y. R., Peterson, R. S., Phillips, D. J., Podolny, J. M., & Ridgeway, C. L. (2012). Introduction to the special issue: Bringing status to the table—Attaining, maintaining, and experiencing status in organizations and markets. Organization Science, 23, 299 –307. http://dx.doi.org/10 .1287/orsc.1110.0668 Chudek, M., Heller, S., Birch, S., & Henrich, J. (2012). Prestige-biased cultural learning: Bystander’s differential attention to potential models influences children’s learning. Evolution and Human Behavior, 33, 46 –56. http://dx.doi.org/10.1016/j.evolhumbehav.2011.05.005 Cooper, C. D. (2005). Just joking around? Employee humor expression as an ingratiatory behavior. The Academy of Management Review, 30, 765–776. http://dx.doi.org/10.5465/AMR.2005.18378877 Cooper, C. (2008). Elucidating the bonds of workplace humor: A relational process model. Human Relations, 61, 1087–1115. http://dx.doi.org/10 .1177/0018726708094861 Costolo, D. (2009, Sept 13). medTwitter. Retrieved from https://twitter .com/dickc/status/3962807808 Daniel, H. J., O’Brien, K. F., McCabe, R. B., & Quinter, V. E. (1985). Values in mate selection: A 1984 campus survey. College Student Journal, 19, 44 –50. Decker, W. H. (1987). Managerial humor and subordinate satisfaction. Social Behavior and Personality, 15, 225–232. http://dx.doi.org/10 .2224/sbp.1987.15.2.225

19

Ellis, L. (1994). Social stratification and socioeconomic inequality: Vol. 2. Reproductive and interpersonal aspects of dominance and status. Westport, CT: Praeger. Feingold, A. (1992). Good-looking people are not what we think. Psychological Bulletin, 111, 304 –341. http://dx.doi.org/10.1037/0033-2909 .111.2.304 Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878 –902. http://dx.doi.org/10.1037/0022-3514 .82.6.878 Flynn, F. J., Reagans, R. E., Amanatullah, E. T., & Ames, D. R. (2006). Helping one’s way to the top: Self-monitors achieve status by helping others and knowing who helps whom. Journal of Personality and Social Psychology, 91, 1123–1137. http://dx.doi.org/10.1037/0022-3514.91.6 .1123 Frank, R. H. (1985). Choosing the right pond: Human behavior and the quest for status. New York, NY: Oxford University Press. Gervais, M., & Wilson, D. S. (2005). The evolution and functions of laughter and humor: A synthetic approach. The Quarterly Review of Biology, 80, 395– 430. http://dx.doi.org/10.1086/498281 Gibbs, R. W. (1986). On the psycholinguistics of sarcasm. Journal of Experimental Psychology: General, 115, 3–15. http://dx.doi.org/10 .1037/0096-3445.115.1.3 Goldhamer, H., & Shils, E. A. (1939). Types of power and status. American Journal of Sociology, 45, 171–182. http://dx.doi.org/10.1086/ 218263 Greengross, G., & Miller, G. (2011). Humor ability reveals intelligence, predicts mating success, and is higher in males. Intelligence, 39, 188 – 192. http://dx.doi.org/10.1016/j.intell.2011.03.006 Halevy, N., Chou, E. Y., Cohen, T. R., & Livingston, R. W. (2012). Status conferral in intergroup social dilemmas: Behavioral antecedents and consequences of prestige and dominance. Journal of Personality and Social Psychology, 102, 351–366. http://dx.doi.org/10.1037/a0025515 Hardy, C. L., & Van Vugt, M. (2006). Nice guys finish first: The competitive altruism hypothesis. Personality and Social Psychology Bulletin, 32, 1402–1413. http://dx.doi.org/10.1177/0146167206291006 Hayes, A. F., & Preacher, K. J. (2014). Statistical mediation analysis with a multicategorical independent variable. British Journal of Mathematical and Statistical Psychology, 67, 451– 470. http://dx.doi.org/10.1111/ bmsp.12028 Hooper, J., Sharpe, D., & Roberts, S. G. B. (2016). Are men funnier than women, or do we just think they are? Translational Issues in Psychological Science, 2, 54 – 62. http://dx.doi.org/10.1037/tps0000064 Hoption, C., Barling, J., & Turner, N. (2013). “It’s not you, it’s me”: Transformational leadership and self-deprecating humor. Leadership & Organization Development Journal, 34, 4 –19. http://dx.doi.org/10.1108/ 01437731311289947 Huang, L., Gino, F., & Galinsky, A. D. (2015). The highest form of intelligence: Sarcasm increases creativity for both expressers and recipients. Organizational Behavior and Human Decision Processes, 131, 162–177. http://dx.doi.org/10.1016/j.obhdp.2015.07.001 IMDb.com. (2015). Office Space quotes. Retrieved from http://www.imdb .com/title/tt0151804/quotes Keltner, D., Capps, L., Kring, A. M., Young, R. C., & Heerey, E. A. (2001). Just teasing: A conceptual analysis and empirical review. Psychological Bulletin, 127, 229–248. http://dx.doi.org/10.1037/0033-2909.127.2.229 Keltner, D., Young, R. C., Heerey, E. A., Oemig, C., & Monarch, N. D. (1998). Teasing in hierarchical and intimate relations. Journal of Personality and Social Psychology, 75, 1231–1247. http://dx.doi.org/10 .1037/0022-3514.75.5.1231 Kennedy, J. A., Anderson, C., & Moore, D. A. (2013). When overconfidence is revealed to others: Testing the status-enhancement theory of

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

20

BITTERLY, BROOKS, AND SCHWEITZER

overconfidence. Organizational Behavior and Human Decision Processes, 122, 266 –279. http://dx.doi.org/10.1016/j.obhdp.2013.08.005 Kilduff, G. J., & Galinsky, A. D. (2013). From the ephemeral to the enduring: How approach-oriented mindsets lead to greater status. Journal of Personality and Social Psychology, 105, 816 – 831. http://dx.doi .org/10.1037/a0033667 Kurtzberg, T. R., Naquin, C. E., & Belkin, L. Y. (2009). Humor as a relationship-building tool in online negotiations. International Journal of Conflict Management, 20, 377–397. http://dx.doi.org/10.1108/ 10444060910991075 Lehmann-Willenbrock, N., & Allen, J. A. (2014). How fun are your meetings? Investigating the relationship between humor patterns in team interactions and team performance. Journal of Applied Psychology, 99, 1278 –1287. http://dx.doi.org/10.1037/a0038083 Loch, C. H., Huberman, B. A., & Stout, S. (2000). Status competition and performance in work groups. Journal of Economic Behavior & Organization, 43, 35–55. http://dx.doi.org/10.1016/S0167-2681(00)00107-4 Lord, R. G., De Vader, C. L., & Alliger, G. M. (1986). A meta-analysis of the relation between personality traits and leadership perceptions: An application of validity generalization procedures. Journal of Applied Psychology, 71, 402– 410. http://dx.doi.org/10.1037/0021-9010.71.3.402 Lyttle, J. (2001). The effectiveness of humor in persuasion: The case of business ethics training. Journal of General Psychology, 128, 206 –216. http://dx.doi.org/10.1080/00221300109598908 Magee, J. C., & Galinsky, A. D. (2008). Social hierarchy: The selfreinforcing nature of power and status. The Academy of Management Annals, 2, 351–398. http://dx.doi.org/10.1080/19416520802211628 Marmot, M. (2004). Status syndrome: How social standing directly affects your health and life expectancy. New York, NY: Macmillan. Martin, R. A. (2007). The psychology of humor: An integrative approach. Burlington, MA: Elsevier Academic Press. Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50, 370 –396. http://dx.doi.org/10.1037/h0054346 Masten, A. S. (1986). Humor and competence in school-aged children. Child Development, 57, 461– 473. http://dx.doi.org/10.2307/1130601 Mazur, A. (1973). A cross-species comparison of status in small established groups. American Sociological Review, 38, 513–530. http://dx.doi .org/10.2307/2094404 McGraw, A. P., & Warren, C. (2010). Benign violations: Making immoral behavior funny. Psychological Science, 21, 1141–1149. http://dx.doi .org/10.1177/0956797610376073 McGraw, A. P., Warren, C., & Kan, C. (2015). Humorous complaining. Journal of Consumer Research, 41, 1153–1171. http://dx.doi.org/10 .1086/678904 McGraw, A. P., Warren, C., Williams, L. E., & Leonard, B. (2012). Too close for comfort, or too far to care? Finding humor in distant tragedies and close mishaps. Psychological Science, 23, 1215–1223. http://dx.doi .org/10.1177/0956797612443831 McGraw, P., & Warner, J. (2014). The humor code: A global search for what makes things funny. New York, NY: Simon & Schuster. Mickes, L., Walker, D. E., Parris, J. L., Mankoff, R., & Christenfeld, N. J. (2012). Who’s funny: Gender stereotypes, humor production, and memory bias. Psychonomic Bulletin & Review, 19, 108 –112. Olson, J. M. (1992). Self-perception of humor: Evidence for discounting and augmentation effects. Journal of Personality and Social Psychology, 62, 369 –377. http://dx.doi.org/10.1037/0022-3514.62.3.369 O’Quin, K., & Aronoff, J. (1981). Humor as a technique of social influence. Social Psychology Quarterly, 44, 349 –357. http://dx.doi.org/10 .2307/3033903 Pettit, N. C., & Lount, R. B., Jr. (2010). Looking down and ramping up: The impact of status differences on effort in intergroup contexts. Journal of Experimental Social Psychology, 46, 9 –20. http://dx.doi.org/10.1016/ j.jesp.2009.08.008

Pettit, N. C., & Sivanathan, N. (2011). The plastic trap self-threat drives credit usage and status consumption. Social Psychological and Personality Science, 2, 146 –153. http://dx.doi.org/10.1177/1948550610385138 Pettit, N. C., & Sivanathan, N. (2012). The eyes and ears of status: How status colors perceptual judgment. Personality and Social Psychology Bulletin, 38, 570 –582. http://dx.doi.org/10.1177/0146167211431166 Pexman, P. M., & Olineck, K. M. (2002). Does sarcasm always sting? Investigating the impact of ironic insults and ironic compliments. Discourse Processes, 33, 199 –217. http://dx.doi.org/10.1207/S153269 50DP3303_1 Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717–731. http://dx.doi .org/10.3758/BF03206553 Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879 – 891. http://dx.doi .org/10.3758/BRM.40.3.879 Provine, R. R. (1992). Contagious laughter: Laughter is a sufficient stimulus for laughs and smiles. Bulletin of the Psychonomic Society, 30, 1– 4. http://dx.doi.org/10.3758/BF03330380 Ridgeway, C. L. (1987). Nonverbal behavior, dominance, and the basis of status in task groups. American Sociological Review, 52, 683– 694. http://dx.doi.org/10.2307/2095603 Ridgeway, C. L. (1991). The social construction of status value: Gender and other nominal characteristics. Social Forces, 70, 367–386. http://dx .doi.org/10.1093/sf/70.2.367 Robert, C., Dunne, T. C., & Iun, J. (2016). The impact of leader humor on subordinate job satisfaction: The crucial role of leader-subordinate relationship quality. Group & Organization Management, 41, 375– 406. Ronson, J. (2015, Feb 12). How one stupid Tweet blew up Justine Sacco’s life. The New York Times. Retrieved from http://www .nytimes.com/2015/02/15/magazine/how-one-stupid-tweet-ruinedjustine-saccos-life.html Rucker, D. D., & Galinsky, A. D. (2008). Desire to acquire: Powerlessness and compensatory consumption. Journal of Consumer Research, 35, 257–267. http://dx.doi.org/10.1086/588569 Sauter, D. A., Eisner, F., Ekman, P., & Scott, S. K. (2010). Cross-cultural recognition of basic emotions through nonverbal emotional vocalizations. PNAS Proceedings of the National Academy of Sciences of the United States of America, 107, 2408 –2412. http://dx.doi.org/10.1073/pnas.0908239106 Sivanathan, N., & Pettit, N. C. (2010). Protecting the self through consumption: Status goods as affirmational commodities. Journal of Experimental Social Psychology, 46, 564 –570. http://dx.doi.org/10.1016/j .jesp.2010.01.006 Smeltzer, L. R., & Leap, T. L. (1988). An analysis of individual reactions to potentially offensive jokes in work settings. Human Relations, 41, 295–304. http://dx.doi.org/10.1177/001872678804100402 Smyth, M. M., & Fuller, R. G. (1972). Effects of group laughter on responses to humorous material. Psychological Reports, 30, 132–134. http://dx.doi.org/10.2466/pr0.1972.30.1.132 Strick, M., Holland, R. W., van Baaren, R. B., & van Knippenberg, A. (2012). Those who laugh are defenseless: How humor breaks resistance to influence. Journal of Experimental Psychology: Applied, 18, 213– 223. http://dx.doi.org/10.1037/a0028534 Thinkexist.com. (2014). Mitch Hedberg quotes. Retrieved from http://think exist.com/quotation/i-had-a-job-interview-at-an-insurance-company/ 365697.html Thomas, C. A., & Esses, V. M. (2004). Individual differences in reactions to sexist humor. Group Processes & Intergroup Relations, 7, 89 –100. http://dx.doi.org/10.1177/1368430204039975 Tiedens, L. Z. (2001). Anger and advancement versus sadness and subjugation: The effect of negative emotion expressions on social status

HUMOR AND STATUS

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

conferral. Journal of Personality and Social Psychology, 80, 86 –94. http://dx.doi.org/10.1037/0022-3514.80.1.86 Veatch, T. C. (1998). A theory of humor. Humor: International Journal of Humor Research, 11, 161–215. http://dx.doi.org/10.1515/humr.1998.11 .2.161 Warren, C., & McGraw, A. P. (2016). Differentiating what is humorous from what is not. Journal of Personality and Social Psychology, 110, 407– 430. Warren, C., & McGraw, A. P. (2015). Opinion: What makes things humorous. PNAS Proceedings of the National Academy of Sciences of the United States of America, 112, 7105–7106.

21

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. http://dx.doi.org/10.1037/0022-3514.54.6.1063 Willer, R. (2009). Groups reward individual sacrifice: The status solution to the collective action problem. American Sociological Review, 74, 23– 43. http://dx.doi.org/10.1177/000312240907400102 Yukl, G. A., Wall, S., & Lepsinger, R. (1990). Preliminary report on the validation of the management practices survey. In K. E. Clark & M. B. Clark (Eds.), Measures of leadership (pp. 223–238). West Orange, NJ: Leadership Library of America.

Appendix A Jokes Used in Studies In Studies 1a and 1b, participants saw another participant (who was actually a confederate) deliver either a humorous or serious testimonial for a hypothetical online pet waste removal service, FastScoop .com (Study 1a), and a hypothetical travel service, VisitSwitzerland .ch (Study 1b). In Studies 2 through 4, participants were presented with a scenario of an interview between a manager and a job candidate. In the scenario, the manager asks the candidate a question. The candidate then responds with either a joke or a serious response.

Serious response: Continuing to work in this field in a role like this one.

Study 3a Manager’s question: Are you looking for a challenging position? Joke response: That’s what she said! Serious response: Yes. I am a hard worker and like challenges.

Study 3b Study 1a Humorous testimonial: Very professional. After cleaning up the poop, they weren’t even upset when they found out that I do not have a pet! But seriously, this service is reliable and always leaves the yard spotless! Serious testimonial: Very professional. This service is reliable and always leaves the yard spotless!

Study 1b Question: What made you fall in love with Switzerland? Humorous testimonial: The mountains are great for skiing and hiking, and the flag is a big plus! Seriously, it’s amazing! Serious testimonial: The mountains are great for skiing and hiking! It’s amazing!

Study 2b Manager’s question: Where do you see yourself in five years? Joke response: Celebrating the fifth year anniversary of you asking me this question.

Manager’s question: What do you see yourself doing within the first 30 days of this job? Joke response: The Receptionist I saw on the way in. Serious response: Getting to know the team and up to speed.

Study 4a Manager’s question: What would you do if you won the lottery? Appropriate joke response: When I die, I would want my last words to be, “I left one million dollars under the. . . .” Inappropriate joke response: I’ll tell you what I’d do, two chicks at the same time. Serious response: I would probably go on a vacation to Hawaii.

Study 4b Manager’s Question: What is a creative use for an old tire? Appropriate joke response: Someone doing crossfit could use it for 30 minutes, then tell you about it forever. Inappropriate joke response: Melt it down, make 365 condoms, and call it a GOODYEAR! Serious response: Make a tire swing out of it.

(Appendices continue)

22

BITTERLY, BROOKS, AND SCHWEITZER

Appendix B

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Sample Stimuli (Study 1a)

(See the online article for the color version of the figures.)

Appendix C Sample Stimuli (Study 1b)

What Made You Fall in Love With Switzerland? (See the online article for the color version of the figure.)

(Appendices continue)

HUMOR AND STATUS

23

Appendix D

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Leadership Election Instructions (Study 1b) At the end of this lab session, you will be asked to engage in a group task with other study participants. In this task, you will complete a team exercise in a small group of 3 to 6 participants and compete against other small groups in this lab session. One person in each group will be the group leader. That person will lead the group in the team exercise. You will elect the group leader by transferring points to each presenter. Every participant has 25 points and has the opportunity to keep some points for him/herself and transfer some points to other presenters. The

person who ends up with the most points will become the group leader. The presenters you just saw may be assigned to your group. You have 25 points. Please indicate how many points you would like to assign to each presenter. The remaining points will be allotted to you. Remember that the person with the most points will become the group leader and will guide your group in the competition, so please answer this question thoughtfully. How many of your 25 points would you like to assign to each presenter?

Appendix E Sample Stimuli (Study 2b)

Additional photos follow with text below: Manager: I’m going to ask a few questions to get to know more about you. Candidate: Sounds good. Manager: Where do you see yourself in five years? In the joke condition, the candidate responds with the following:

Candidate: Celebrating the fifth year anniversary of you asking me this question. And participants then read: “The manager and candidate both laugh.” The photos shown above are very similar to the ones used in the study (the originals are available upon request from the authors).

(Appendices continue)

24

BITTERLY, BROOKS, AND SCHWEITZER

Appendix F Mediation Analyses for Studies 1 Through 4 We conducted both Baron and Kenny (1986) and bootstrap analyses (Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008) to test for mediation. We report the Baron and Kenny (1986) analysis below and report the results of the bootstrap analysis in Table 5.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Study 1a Perceptions of confidence and competence mediated the relationship between the second presenter’s testimonial and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included competence ratings for the second presenter in our model, with status and competence ratings for the first presenter as covariates, the effect of the condition was reduced (from ␤ ⫽ .67, p ⬍ .0001 to ␤ ⫽ .43, p ⬍ .01), and the effect of competence remained significant (␤ ⫽ .41, p ⬍ .0001). When we included confidence ratings for the second presenter in our model, with status and confidence ratings for the first presenter as covariates, the effect of the condition was reduced (from ␤ ⫽ .60, p ⬍ .0001 to ␤ ⫽ .18, p ⫽ .08), and the effect of confidence remained significant (␤ ⫽ .39, p ⬍ .0001).

Study 1b Perceptions of competence mediated the relationship between the second presenter’s testimonial and our behavioral and attitudinal measures of status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). Perceptions of confidence mediated the relationship between the second presenter’s testimonial and our attitudinal, but not our behavioral, measure of status. Behavioral measure of status. For the allocation of leader points to the second presenter, our behavioral measure of status, when we included competence ratings of the second presenter in our model, with leader points allocated to the first presenter and competence ratings of the first presenter as covariates, the effect of the condition was reduced (from ␤ ⫽ 1.64, p ⬍ .01 to ␤ ⫽ 1.31, p ⬍ .05), and the effect of competence was marginally significant (␤ ⫽ 1.13, p ⫽ .07). When we included confidence ratings of the second presenter in our model, with leader points allocated to the first presenter and confidence ratings of the first presenter as covariates, the effect of the condition was reduced (from ␤ ⫽ 1.38, p ⬍ .05 to ␤ ⫽ 0.96, p ⫽ .14), and the effect of confidence was not significant (␤ ⫽ 0.50, p ⫽ .30). Attitudinal measure of status. We next consider attitudinal ratings of status of the second presenter. When we include the

competence ratings of the second presenter in our model, with status and competence ratings of the first presenter as covariates, the effect of the condition was reduced (from ␤ ⫽ .38, p ⬍ .001 to ␤ ⫽ .17, p ⬍ .01), and the effect of competence remained significant (␤ ⫽ .60, p ⬍ .0001). When we included confidence ratings of the second presenter in our model, with status and confidence ratings of the first presenter as covariates, the effect of the condition was reduced (from ␤ ⫽ 0.38, p ⬍ .01 to ␤ ⫽ 0.06, p ⫽ .48), and the effect of confidence remained significant (␤ ⫽ 0.37, p ⬍ .001).

Study 2a Affect did not significantly mediate the relationship between a successful humor attempt and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included positive affect in our model, included negative affect as a covariate, and compared the Successful Joke condition with the Serious Comment condition, the effect of the Successful Joke condition remained significant and was only slightly reduced (from ␤ ⫽ .77, p ⬍ .05 to ␤ ⫽ .68, p ⬍ .05), and the effect of positive affect remained significant (␤ ⫽ 0.35, p ⬍ .001). When we included negative affect in our model, included positive affect as a covariate, and compared the Successful Joke condition with the Serious Comment condition, the effect of the Successful Joke condition remained significant and was very slightly reduced (from ␤ ⫽ .678, p ⬍ .05 to ␤ ⫽ .677, p ⬍ .05), and the effect of negative affect was not significant (␤ ⫽ ⫺0.02, p ⫽ .82).

Study 2b Perceptions of confidence and competence mediated the relationship between the successful use of humor and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included competence in our model and compared the Successful Joke condition with the Serious Response condition, the effect of the Successful Joke condition was reduced (from ␤ ⫽ .79, p ⬍ .0001 to ␤ ⫽ .43, p ⬍ .0001) and the effect of competence remained significant (␤ ⫽ .76, p ⬍ .0001). When we included confidence in our model and compared the Successful Joke condition with the Serious Response condition, the effect of the Successful Joke condition was no longer significant (from ␤ ⫽ .79, p ⬍ .0001 to ␤ ⫽ .06, p ⫽ .65) and the effect of confidence remained significant (␤ ⫽ .54, p ⬍ .0001).

(Appendices continue)

HUMOR AND STATUS

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Study 3a Perceptions of competence fully mediated the relationship between the failed use of humor and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included competence in our model and compared the Failed Joke condition to the Serious Response condition, the effect of the Failed Joke condition was no longer significant (from ␤ ⫽ ⫺1.27, p ⬍ .0001 to ␤ ⫽ ⫺.11, p ⫽ .37) and the effect of competence remained significant (␤ ⫽ .64, p ⬍ .0001). We also tested whether perceptions of confidence mediated the relationship between humor and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included confidence in our model and compared the Failed Joke condition to the Serious Response condition, the effect of the Failed Joke condition increased (from ␤ ⫽ ⫺1.27, p ⬍ .0001 to ␤ ⫽ ⫺1.35, p ⬍ .0001) and the effect of confidence remained significant (␤ ⫽ .37, p ⬍ .001).

Study 3b Perceptions of competence fully mediated the relationship between the failed joke and decreased status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included competence in our model and compared the Failed Joke condition with the Serious Response condition, the effect of the Failed Joke condition was significantly reduced (from ␤ ⫽ ⫺.93, p ⬍ .0001 to ␤ ⫽ 0.00, p ⫽ .97) and the effect of competence remained significant (␤ ⫽ .59, p ⬍ .0001). We also tested whether confidence mediated the relationship between the failed joke and decreased status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included confidence in our model and compared the Failed Joke condition with the Serious Response condition, the effect of the Failed Joke condition increased (from ␤ ⫽ ⫺.93, p ⬍ .0001 to ␤ ⫽ ⫺1.21, p ⬍ .0001) and the effect of confidence remained significant (␤ ⫽ .38, p ⬍ .0001).

Study 4a Perceptions of confidence and competence fully mediated the relationship between the Appropriate Successful Joke condition and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included competence in

25

our model and compared the Appropriate Successful Joke condition with the Serious Response condition, the effect of the Appropriate Successful Joke condition was no longer significant (from ␤ ⫽ .55, p ⬍ .01 to ␤ ⫽ .25, p ⫽ .10), and the effect of competence remained significant (␤ ⫽ .61, p ⬍ .0001). When we included confidence in our model and compared the Appropriate Successful Joke condition with the Serious Response condition, the effect of the Appropriate Successful Joke condition was no longer significant (from ␤ ⫽ .55, p ⬍ .01 to ␤ ⫽ .11, p ⫽ .57), and the effect of confidence remained significant (␤ ⫽ .29, p ⬍ .0001). We also tested whether confidence and competence mediated the relationship between the Inappropriate Failed Joke condition and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included competence in our model and compared the Inappropriate Failed Joke condition with the Serious Response condition, the effect of the Inappropriate Failed Joke condition was no longer significant (from ␤ ⫽ ⫺1.11, p ⬍ .0001 to ␤ ⫽ ⫺.13, p ⫽ .46). When we included confidence in our model and compared the Inappropriate Failed Joke condition with the Serious Response condition, the effect of the Inappropriate Failed Joke condition increased (from ␤ ⫽ ⫺1.11, p ⬍ .0001 to ␤ ⫽ ⫺1.40, p ⬍ .0001).

Study 4b Affect did not mediate the relationship between an appropriate humor attempt that succeeds and status (Baron & Kenny, 1986; Hayes & Preacher, 2014; Preacher & Hayes, 2004, 2008). When we included positive affect in our model, included negative affect as a covariate, and compared the Appropriate Successful Joke condition with the Serious Response condition, the effect of the Appropriate Successful Joke condition remained significant (from ␤ ⫽ .90, p ⬍ .0001 to ␤ ⫽ .92, p ⬍ .0001) and the effect of positive affect remained significant (␤ ⫽ .31, p ⬍ .0001). When we included negative affect in our model, positive affect as a covariate, and compared the Inappropriate Failed Joke condition with the Serious Response condition, the effect of the Inappropriate Failed Joke condition remained significant (from ␤ ⫽ ⫺1.37, p ⬍ .0001 to ␤ ⫽ ⫺1.29, p ⬍ .0001) and the effect of negative affect was not significant (␤ ⫽ .05, p ⫽ .41). Received September 4, 2015 Revision received August 24, 2016 Accepted August 26, 2016 䡲